Paper I

Table of Contents Introduction 3 Need for technology-based solutions 3 Infrastructure Automation Tools 4 Implementation 4 The Central Theory: Organizational Management and Memory 4 Organizational Management 4 Organizational Memory 4 Need of Data Archival And Storage 5 Data Storage. 5 Types of Storage. 6 Data Archival 9 Data Archival Process 9 Archiving principles 12 Data Management Systems 12 Enterprise Resource Planning Systems (ERP systems) for data integration. 13 Microservices. 15 Properties of Monolithic. 17 Conclusion 22 References 24

Technology is considered vital in today's globalized world. Especially in terms of business, information technology has both quantifiable and unquantifiable benefits. It is essential to communicate with customers and stakeholders regularly and necessary for communicating quickly and clearly. It helps in implementing business operations efficiently and effectively, also. A business with robust technological capacity creates new opportunities for a company to stay ahead of the competition and grow eventually (Rangus & Slavec, 2017). Consequently, it also makes dynamic teams that can interact from anywhere in the world—furthermore, technology aids in understanding the business needs and managing and securing confidential and critical data.

The need for data recovery, active and continuous data processing by its life cycle of significance and utility for research, scientific and educational purposes (Bukari Zakaria & Mamman, 2014). The acknowledgment that information is an organization's key asset since late, decisively affecting its profitability, has contributed to some comprehensive corporate memory approaches. The key causes of competitive advantage are corporate memory and organizational learning ability (C. Priya, 2011). Hence the main obstacle is the effectiveness of information management while ensuring the consistency of training facilities.

Organizations need robust technology-based solutions. Thus, software developers have developed and deployed various forms of overtime architectures that enable software products to become resource-effective and usable. Some architectures implement their frameworks in either one layer or various layers or levels (Suresh, 2012). It is understood that ERP implementation efficiency of ERP implementations is influenced by the rise or excess of a certain degree of capability in the volume of data to process (Johansson, 2012). In the last couple of decades, new architectures have been created with creativity that offers optimum solutions. Thus, the microservices architecture is gaining room and becoming part of the technological, financial, and advertising decision-making process. Microservices replace monolithic, tightly dispersed system-focused applications with an independent operation (Vrîncianu, Anica-Popa, & Anica-Popa, 2009).

One issue as microservices are applied is that any service operation must be implemented and measured in the cloud. Companies deploying microservices can also use various automation platforms such as DevOps, Docker, Chef, Puppet, and automated scaling. These instruments save time and money by implementing them (Balalaie et al., 2018).

Regrettably, further growth, migration, and integration are required. Thus, infrastructure costs are the key focus for companies adopting the listed trends to achieve agility, autonomous development, and scalability. Another challenge is the output ensemble of microservices. While it could solve any apparent technological problem, its configuration and capabilities must still be consistent with the new architecture. Though different solutions still exist, there is still no precision assessment of transitioning from ERP architectures to microservices.

In one study proposed by Stubbs et al. to discuss container technologies in microservices design and service exploration difficulty. The authors propose, based on the Serf initiative, a decentralized open-source approach. They defined the construction of a synchronization solution of data files between repositories using Docker using Git. Due to this report's findings, Serfnode was identified, which unites the Docker's containers with another community of existing clusters and does not impact the original container's dignity (Stubbs et al., 2015).

Similarly, the approach allowed frameworks for control and oversight that perfectly completed the container because they allow the applications operating in each shared space to be isolated and independent. While containers can simplify containers' use and delivery, they do nothing to solve underserviced connectivity through a complex network. Finally, this research examines alternatives that allow Microservices and Containers to be used to the greatest possible extent.

In terms of implementing the above, according to Sandoe & Olfman, corporate memory is in line with IT advances and can counter much unnecessary organizational forgetfulness. The paper shows how structuring philosophy can be used to bridge irreconcilable views (Ehrhart et al., 2015). The paradigm presented in this paper shows that collective memory comprises laws and tools that remedy interactivity and organizational structure. This model is appropriate for the categorization of current and future co-memory structures based on IT. Comprehensively, the paper's forecast shows a mnemonic transition in culture to discursive organization models that primarily depend on IT-based co-membrane (Sandoe & Olfman, 1992).

The contrast between microservices implementations to ERP architecture has been clarified by Singh & K Peddoju. These authors deployed the proposed Docker Container Microservices and tested them as a case study utilizing a social networking framework. Because of the efficiency contrast, Jmeter8 was built and used to apply constant applications for both designs. For the design of the ERP, the application has been forwarded with a web-based API. By comparison, HAProxy is used to send queries to the intended service for a microservices architecture. The findings showed that the application designed and implemented using the microservices method decreases the time and commitment required for the application to be deployed and continually integrated. Their findings have also established that the ERP paradigm is superimposed due to low response times and good performance by microservices. Our experimental findings show that containers are acceptable launches compared to virtual machines for microservices applications (VMs). Several suggested experiments have been conducted on the benefits and drawbacks of moving from an ERP to one of the microservices architecture (Singh & K Peddoju, 2017).

Sandoe and Olfman (1992) and Morrison (1997) describe two organizational management forms that satisfy two functions: representation and interpretation. Representation presents the circumstances for a given situation or position. Analysis promotes adaptation and learning by offering frames of character reference, methods, regulations, or a means to synthesize past information for application to new situations (Organizational Memory). This theory is especially applicable in using information systems. Organizational and cultural factors play a major role in the optimal functioning of information systems (Booth & Rowlinson, 2006). Specifically, the implementation of robust services needs well-defined contracts with all teams involved rather than catering to each team's individual/special needs. Organizational dynamics determine how the contracts, as mentioned above, are negotiated, designed, and implemented.

Organizational dynamics are rooted in an organizational culture defined as patterns of shared values, beliefs, and assumptions underlying behavioral norms between organizational members (Schein 1992). This definition implies that the culture is persistent and rooted in the shared history and experiences developed over a long time. Hence, organizational culture plays a long-term role because this cultural persistence has become important in understanding resistance to new IT implementations and their subsequent adoption. In global organizations, national sentiments expand the scope of organizational culture.

Empirical knowledge is a key to competitiveness. Therefore, conservation of organizational memory is growing progressively essential to organizations. With the convenience of innovative information technologies, information systems become a crucial part of this memory (Perez & Ramos, 2013).

Organizational Memory Information Systems (OMIS) bring together culture, history, business process, human memory, and the actuality into an integrated knowledge-based business system. OMIS's assist businesses in fitting in different databases, capturing the skill of retiring staff, enhancing organizational expertise, and providing decision-making support to employees facing new and complex issues while integrating disparate and uneven types of knowledge (Roth & Kleiner, 1998).

The organizational memory dictated by culture is continuously exposed to restructuring and change, is recreated, reconfigured, and enhanced by new knowledge by organizational learning procedures through shaping organizational performance by capitalizing and evaluating the cognitive acquis of the enterprise (Linger et al., 1999). The company is most frequently described as an "elaborate, immaterial and permanent representation of knowledge and facts." The organizational memory diagrams an organization's cognitive infrastructure that enables an organization to recognize, compile, convert, capitalize, and value awareness, facts, rules, and community values.

Certain analysts have evaluated almost 40% of the Fortune 500 firms' activities in 2005 as part of their corporate learning, using some form of information management systems (Siong Choy & Yong Suk, 2005). This study exposed some critical aspects of organizational culture that reduce the efficiency of information management systems.

Too frequently, when preparing digital workspace programs, digital archive projects are set down on a priority list. Business is incorrect to assume that low storage expenses and a powerful search engine require all its records. What would go wrong, and besides? For knowledge processing, archiving is crucial and will allow a company more oversight of their data operations. When an organization expands, more data is generated – data that need closely handled and controlled to be correctly used. Holding tabs on these records can be difficult for firms who never implement an archiving scheme (Borgerud & Borglund, 2020). Records not archived becomes harder to find, protect and distribute when housed in a surrounding environment - like a desktop - and would thus be useless to other user groups. This will potentially adversely impact organizational operations and the morale of workers.

The main purpose of data collection is to digitally archive files and records and preserve them for the storage facility's potential use. If required, storage systems may depend on electromagnetic, mechanical, or other devices to conserve and restore data. Data storage allows it to archive files in case of an accidental computer crash, data breach, or data archiving for safekeeping and fast recovery (Spoorthy et al., 2014).

Although not all databases must be preserved, it is necessary to preserve what needs to be preserved to make the data safe and accessible. Data storage refers to a variety of ways in which physical media store information to be accessed once users require it (XIE & CHEN, 2013). In the evolution of computation, the storeroom equipment has greatly evolved over the years, from room size microprocessor computers' electromagnetic instruments to state-of-the-art solid-state drive technologies (SSDs) and, just like many products in the technical field, these approaches keep evolving as the need for data and storage increases.

Data storing on physical hard discs, discs or USB drives or in the cloud is possible. The main thing is, if your machine ever crashes beyond recovery, the files are substantiated and readily accessible. Reliability, the strength of security capabilities and the costs of implementing and maintaining the infrastructure are among the most critical things to remember when it comes to data storage (Esposito, 2018). By browsing various data storage systems and products, one can make the most suitable option for your enterprise.

The corporation's storage style plays a major role in the accessibility of its records, the number of archive expenses, and the data's safety after it has been archived. An archive is only valuable when data can be accessed when necessary, so it should be regularly checked that the stock chosen by the organization is still working.

Offline storage.

Data undoubtedly grow, but one of the traditional storage types still has a role in modern industry. Offline backup has been there for years and includes archiving vital files using digital discs such as CDs and Blu-Rays. And if the data is not accessible immediately since the storage choice is more new, offline storage is extremely protected while being accessible in the event of a network outage.

The offline storage is also ideal if the company has regulatory obligations or if knowledge for legal purposes has to be supplied. It should be maintained on a written media to ensure the information is lawfully permitted. RAID discs and other cloud storage cannot be placed (Chan Jianli et al., 2020).

Online storage.

Although it can sound intuitive to include all online storage in a similar classification, two distinct offers currently exist. Then the online storing facilitates the store of data stored in the cloud for customers and companies. This is what researchers mean by cloud storage for the objectives of this post. Cloud storage will function very well, provided it progressively safeguards data and does not require upfront resources (Rausher et al., 2010). The drawback, however, is that it could be unacceptable for data to be collected if complete data retrieval is required.

Any businesses that have taken a cryptographic signature of cloud service to develop some of the advantages of energy and reliability are not happy putting their information in the hands of 3rd Party cloud infrastructure suppliers. While this was when out of small enterprises' grasp, advances now enable small enterprises to tap into personal cloud storage.

Cold Storage.

Data less commonly viewed and does not, therefore, require fast access to colder data. This contains information that is no longer being used actively and may not be necessary for months, years, centuries, or even ever. Practical forms of cold-storage documents include ancient ventures, information used to hold other company records, or something worthwhile but not needed shortly (Zhao et al., 2020). Data recovery and reaction times are usually much longer than those for the active control of data on Cold Cloud storage networks. Services such as Amazon Glacier and Google Coldline are practical instances of cold cloud computing (Zhao et al., 2020).

Cloud Storage

Cloud storage is the organization, with the required rights, of data stored anywhere that everyone can reach on the Internet. You do not have to be wired to a corporate network because you do not have access to information on devices. Microsoft, Google, and IBM are common cloud storage providers (Yuhuan, 2017). Cloud storage is supported by cloud-based IT ecosystems that allow cloud computing to operate cloud-based tasks. Cloud storage has no internal network access or specific data storage connectivity.

Data services are differentiated from hardware devices as the basis of a cloud storage volume. Network virtualization is one approach to dissect, taking a dozen separate servers (either convenient or confidential) and abstracting computing capacity. This entire virtual storage area can be grouped into an information lake termed a unified repository, accessible to consumers (Langos & Giancaspro, 2015). That generated cloud storage when such information lakes are linked to the web.

Block storage.

The storage block is also designed to separate the user interface information and be best used in various contexts. However, when information is analyzed, the storage program reorganizes and returns the information blocks from those contexts. It is normally used in SAN settings and has to be connected to a working server (Kumari et al., 2019). This must be done on a network.

The storage process cannot be easily retrieved because it would not consist of a single physical layer, like files' storage. The blocks are separate and can be subdivided to enable them to be accessible from another web browser, enabling them to customize their data. It is an inexpensive, secure, and user-friendly way of storing data (Fujita & Ogawara, 2005). It goes best for companies that carry out large transactions and introduce massive databases, suggesting that the more information they have to store, the easier that can get with block storage.

However, there are a few downsides. The storage of blocks can be costly. It has no metadata handling, meaning that it must be handled based on a program or database—adding something else to think for a programmer or server operator.

Data archiving is a practice in which data that is not operational anymore are identified and transferred from processing to long-term storage systems. Archival files are preserved in order to be able to be returned to service at any point. Archived records are processed at a lower cost level to reduce primary disc consumption and associated costs. A significant part of a company's data archiving policy is archiving the data and classifying data as an archiving nominee (McDaniel, 2014).

Purpose.

Businesses will store data for business image objects through a data archiving mechanism. This method is carried out by a business process-related archiving entity, though, in this file storage subject, the arrangement or arrangement of the data is specified. When the data are archived, the machine copies the information to archive archives scans the archived data after multiple tests, and, if accurate, extracts it from the operating system. In contrast to the main method, subfields for viewing and reloading archived data and device profiles still exist (Hujda et al., 2016).

Preparing the data.

As a source of information, the company has all aspects of its software project (files, resources, source files, test reports, etc.). (SVS). Consequently, the setup is checked to ensure that none is lacking. There is no problem. Till all the elements accessible are checked, an archive may be created. It must be a robust database, and companies must set the time of archiving. The period of the archiving is contractually, contextually, and risk-based. The archiving media and procedure have to be modified according to the period. Verification is needed for archiving on external drives (Kornei, 2019). For archiving on an external hard drive, a validation procedure is essential, and discs are changed regularly.

Process Flow.

Major comment threads, including the study, writing, and deletion, form the fundamental archiving process. You may combine these if the appropriate customization settings are made. Parallel systems for research and writing may be handled if the parallel analysis method is used. To accomplish this, appropriate data packages are created, which are parallelly processed by separate jobs. The subprocess initially analyses the archiving object data set and then creates the appropriate package templates for parallel processing specified by the program cap (Bruno, 2014). Profiles are insistently saved in the archive and subsequently used by the research and writing subcategories if configured for the archival object in the global personalization settings.

To minimize the overarching runtime of the archive project, the software profile creation aims to use simultaneous package managers to review and write subprocess. The dataset must, however, further than practicable, be split into different packages of the same size (Ribeiro, 2001). As the data distribution can alter over time, the pre-step needs to be repeated regularly to ensure that appropriate profiles are provided.

Simulation.

The simulation feature excluding the deletion or labeling of commercial items from the operating database would follow all the archive procedure phases. It is just a test run. In real fact, it generates an archive destination address, which differs between it and figures. This could be utilized to check the predicted performance (Onggo & Hill, 2014). Although you could use a test database other than the true operating bases to do most of the same research, the benefit of using the actual thing makes it a better test. It guarantees that the conditions for the evaluation are operationally consistent with the database specification.

Write.

The analytical software begins immediately with the normal setup. The writing process clones the specified information from the operating database to files in the research sub-process. Consequently, the information is archived throughout this process.

Like in the research method, the data are recorded in parallel, discrete-time positions. Each task processes the various sub-packages from the collective file. With each parallel processing task, exactly one archive is formed; it can comprise one or even more documents.

Delete.

The erase subprocess extracts data from the operating storage as the data is copied into the backup archives. To do so, stored records can be accessed and removed only if they are read from the archive effectively. This protocol ensures that unless the machine is equipped to archive data according to the guidelines and configurations, data is deleted from the database. A regular setup begins automatically when the write operation finishes a single archive file successfully. The amount of generated deletion procedures is often the same as the number of documents generated by the written program.

If an archive document ceases to be accessed, the data is left to be archived in the operating system and is retrieved again during the next archive exercise by the writing process. Either you should selectively remove the already generated archive files or keep them in the archive. The latter choice is innocuous because files will not be deleted from the OS: Only when an erase process has been accomplished effectively are archive information systems developed.

Data integrity issues.

When data is archived, it is often usually removed from the database from which it is archived. If replicates of these data are present in other databases, data combinations could not be compatible with these data. When all records are made, the results can vary. This can lead to an alarming condition where most consumers in one system vary from those of other system users (Khidzir & Ahmed, 2018). It might even be crucial to provide an extended archiving method, which deletes copies from other records simultaneously if databases are to maintain continuity.

Accessing the data.

The data is stored in a separate archive pool until the project is completed. It is different from every file stream that is generated for the substitution program. Access to the pool of retired applications should be independent of exposure to the archive's water stream. Perhaps both would not fit with the access rationale. This increases to a degree above the logic of breaking metadata. The developed archive channel will verify the 100% rule for all archive access criteria (Senko, 1977). Once the archive has been completed, the source request will be lost. It saves much money. There is no way to re-examine data in the source networks. This ensures that users must perform rigorous access tests before anyone can claim success.

Data archiving is the method of preserving the activities for further scanning and review within the framework. When information passes into the repository for the processing device, an archive file collects and saves data in an indexed fashion for recuperation. Data is normally saved for alarm/access regulation, adjusting device status, streaming, and audio. It is probably to be housed on individual disc or library volumes (Vans et al., 2018).

In the management of their archives, archivists implement the two concepts of 'provenance' and original order. These ideals must form the basis for all the archives' practices (Kilchenmann et al., 2019). Until any take action to enhance their maintenance and care, your archives ought to consider how and how they were made and how well they are organized.

Original Order.

In the sequence in that, they were first produced or used; archives are stored. 4 This must be understood when dealing with libraries to maintain this original order. The original order enables guardians to safeguard the validity of documents and contains important knowledge on the type, maintenance, and usage of records. Perhaps this initial order was missing due to misuse or "re-sorting" (Stokes, 2012).

The original order essentially ensures that objects remain in the order that the individual or organisation whose archives initially held them. This is significant even though those documents might be stored intact for a purpose, even though the purpose was not readily evident.

A basic concept of archive management is a consideration for the initial registration order. Digital archive arrangements are far less about preserving the actual structure of storage media but are more about retaining logical links between electronic records when the digital records' external order frequently requires to be changed for storage and maintenance purposes (Niu, 2014).

Provenance.

The provenance theory ensures that the documents that a person or organization creates accumulate and maintain collectively to be separate from some other maker's documents. As its development promoted the challenges caused by archival science, the concept of origin is regarded as a landmark in archival practice and philosophy (Milosch, 2014).

Provenance signifies the history of possession of the holding of a set of documents or an object. This implies the designers and later owners of the documents and their relationship with the files. It is important to preserve knowledge about those partnerships because they show how and who produced and used the documents before becoming something of the archive. Provenance offers important historical material to appreciate the contents and heritage of a series of archives (Hunter & Cheung, 2007).

As the notion of origin originated in an archival sense in the 19th century, it had a logical objective: to arrange a collection of documents that had lost their organic association with its authors as a result of a thematic grouping. This theory led the archivists to apply the theory as a concrete organizational principle, which consolidates archives of the same love. One reason documents cannot be lent is provenance. The possession and retention (physical presence and not content) of an archive after it was established should preferably trace (Tognoli & Guimarães, 2018). The shareholders' knowledge lets one assess whether anyone has modified it, so it is easier to say whether it would be genuine.

Archival Locations.

When we intend to archive records, we must worry about disaster recovery and enterprise continuity plans, which can turn very difficult because the archiving process recognizes threats. Let us presume we want to archive records; it is normally a terrible idea to store archival information in the same space or building as the facility used for data retention (Leonhardt et al., 2016). We determine that the archived intrusion test data should be maintained in a safe facility that is physically separate from the system site, so natural and human-made accidents are ever at risk. We need two versions – one centrally and one else if we need it fast.

Compliance.

Due to legal enforcement, certain organizations are forced to maintain data for a specified amount of time. It is a prominent market issue that remains under regulatory criteria as required by industrial laws or governmental policies. Consequences can comprise payments for costs, fines, and canceled contracts for breach of compliance (Giacalone et al., 2018).

Data archiving allows companies to achieve compliance through long-term data storage as well as consolidation in an audit. The rules governing the time required to store, store and have access to information vary depending on the sector and the form of data companies generated in this industry. The below are among some of the explanations why organizations focus on methods for data archiving:

Preventing data loss.

For legal purposes, archiving is also relevant. Many corporations have records that really should be kept by regulation unintentionally. Thus, staff should remember that breaching such rules will lead to heavy sanctions or even imprisonment punishments in certain contexts. The movement of data was one of the most serious challenges to ongoing implementations and data access. Statistics suggest that this year there will be an unplanned collapse in 75% of organizations (Killalea, 2016). When an organization executes a data migration process, there is a much greater risk of failure. Archiving preserves business processes and the company's data by transferring data from costly main storage facilities to a significantly lower-cost archive storage device.

Legal requirements.

A successful archiving scheme guarantees consistency with company-specific retention schemes, independent of individual workers' expertise. To make it conscious that violating these rules could contribute to substantial fines or prison terms in some situations (Gerber & von Solms, 2008), the Data Protection regulators are imposing more strict penalties on the industry.

When an entity deals in a court action, the company is provided with encrypting protection and assistance. This is generally termed a discovery in lawsuits and is data collection and transmission on request (S & Venkateshkumar, 2018). Excluding archives, the expenses of gathering evidence for a complaint could be as cost-effective as the case itself.

Data Backup Optimization.

Data backups can be slow and tedious, but it does not have to be this way when they store the corporate data. Indeed, certain businesses that archive files see significant changes in data retention times and, for this kind of purpose, some also switch to file archiving. Choosing file archiving firms is one step forward by supplying archive duplication to remove the necessity for data backups. This is much more cost-effective and productive (Ghantasala et al., 2018).

Data Storage Costs.

Perhaps this one is more evident. Info, period, is paid for the business. Regardless of the business or data form, it costs a fortune to retain the data on a disc or a cloud. It does not matter which one is (Sergeant & Sergeant, 2010). It is the financial part that has the greatest advantage to archive the results. The cost will be minimized by up to 50 percent based on your data amount as users store the business's data. This will contribute to considerable long-term savings for products and other sectors.

Data security and compliance.

The purpose of the data protection conformance regulations is to enable businesses to ensure that data structures and sensitive data are integral, secure, and available. They have a series of protocols and regulations that safeguard companies from security vulnerabilities by protecting networks and records (Bindley, 2019).

Controlled companies are accountable for maintaining records rather than frequently(Sholler et al., 2019). This is to follow regulations as well as the principles of conformity. Regardless of its scale or sector in which it works, conformity can be seen in virtually every record made. In comparison to GDPR, archiving is a prerequisite if the organization complies with all other organization legislation, particularly data management.

Data Storage Management.

Maybe the most prominent purpose for archiving data is for efficiency and sizing purposes to eliminate redundant transaction data in the output record (Sangat et al., 2017). The data provider should also point to the data kept in its primary storage.

While it seems to be a minor act, the organization will save both main and backup storage IT expenses and improve the speed of software such as IFS Applications. In exchange, a quicker machine can boost efficiency. Finally, the company has to hold up its space on its main datastore but does not address it anymore by correctly archiving obsolete data firms.

Data Visualization.

The user can help digest the data with visualization and view new directions. This allows consumers to recognize new dynamics and phenomena that may not be seen with table data. It enables managers in graphical displays, including diagrams, plots, and heat maps, to monitor results (LaPolla & Rubin, 2018).

As big data emerges, data visualization is increasingly essential to interpret the data gathered daily by the data user.

Data visualization enables companies with modern, more immersive formats to recognize, analyze and communicate data. This willingness to be data-oriented allows them to educate and learn how to use data visualization applications and their related formats. The best data archiving strategies enable companies to visualize their data and create better strategic strategies for their records (S & Sathayanarayana, 2018). It ensures that you can understand just how aged the database is and what data the firm has, how many days the data is processed, and other important information that allows the company to build effective data archiving policy.

Increased Security.

Maintaining outdated or inactive records on high-traffic databases raises the prospect of a possible intrusion on the enterprise and with correct access controls. By protecting unuseful data, for instance, by separating them from public access to a remote backup tier or system, businesses reduce the risk and possible effect of data lost or stolen while ensuring the confidentiality of all these data as long as they are required (Schafer, 2004).

For security purposes, archiving is critical, particularly when cyber-attacks and data violations are getting more prevalent. Companies can detect and defend themselves from unwanted third parties by safely archiving records.

Data Consolidation.

The organization wants to optimize this information and so much data through its computers, which expands rapidly every day. De-duplicate and stubborn files are just the start! Any file archiving program helps you to further certain compact files, lessening the digital footprint (Narayanan, 2020). If this is not an opportunity to archive information for the organization, what is it?

Companies can obtain major information efficiently and conveniently by data consolidation. Businesses may improve their production and competitiveness if valuable knowledge is saved in a single location. Data restructuring also decreases maintenance costs. From either the context of data intake, the solution to the sustaining data illustration is more complicated with a larger number of references being incorporated into the Key framework (Bergquist, 2001).

Data is a distinct piece of information, usually formatted in a certain manner based on user requirements. Data should be stored and archived in structured and encrypted form for traceable and secure access. Data storage, archival, and data retention are critical to the organization for both business and legal reasons (Bose, 2006). Lack of good practices in an organization can open and organization and its employees to several risks, which could damage an organization's reputation and business. For example, in the health industry, data safety and patient confidentiality are paramount (Ferrari, 2010).

Archiving data ensures robust backup, faster recovery of data guarantees easier backup processes. It also helps in maintaining and protecting the policies and objectives of and organizations and less time-consuming. An efficient data storage pipeline/strategy and archival and cost-effective archival solutions enhance productivity and lead to organizational growth.

Information systems in a business can be composed of custom applications (written internally) or commercially purchased generic systems. Custom applications require extensive resources, long and expensive development cycles. Moreover, they need to be continually updated and maintained with the evolving landscape of new information systems (Herrmann, 2016). Off-the-shelf commercial systems remove the above problem by taking the responsibility off the user. However, the one size fits all generic commercial systems approach cannot be tailored specifically to each business requirement (which may have thousands of parameters), thereby imposing the need to obtain IT solutions from several different vendors (Wickramasinghe & Gunawardena, 2010). Separate modules are needed to link different functional areas. For example, the human resource area will require a different module to satisfy its business needs compared to the financial area.

Data generation and handling.

These modules should be linked to make better business decisions by using the data generated from each module across each other. Enterprise resource planning (ERP) systems were developed with this vision. ERP applications are implemented to provide an integrated solution to all areas involved in the business operations (for example, Human resources, sales, etc.). ERP applications are solely developed for data handling and are thus well suited for modeling various transactional processes (Pylypenko & Redko, 2019). These systems consist of applications focused on the integration of data from various sources. Common data structures are shared across many applications and thus eliminate the need to pass data step-by-step among other applications. In ERP systems, data manipulation is easy since data is maintained in interoperable databases that can store data in a structured format used by the ERP applications. This, in turn, is based on the assumption that data infrastructures are homogeneous across the organization (rarely the case), which means that in some cases, databases are from the same vendor.

Moreover, some ERP systems may only support databases from a specific vendor, forcing them to adopt standardized data management solutions according to the ERP system. This also means that the adoption of specific ERP systems requires that legacy databases be replaced with ERP–compatible databases, which creates the need for data conversions and the creation of defined architecture for data storage. Therefore, the conversion from legacy databases to ERP-compatible versions needs standardizing, transferring, and cleaning existing data elements (Lee & Chang, 2020).

To ensure its effectiveness and stability, the construction of the ERP Mechanism plays an important part. Three essential architectures for ERP systems are currently established (figure below). The architecture utilizes a specific technology for implementation to restrict the use of an appropriate method for any role the device has to perform. The threats of short- or long-term issues are not well understood.

Stable structures, the complete management structure developed by leading organizations, including IBM, Sun Microsystems, and BMC, are among the most important advantages (Khazaei et al., 2016). These technical providers provide a high degree of product expertise. Some drawbacks, such as hierarchical structures and new requirements.

The improvement in computing power means the existing server is transformed into a bigger server. Its equipment is patented, and it makes the retailer reliant on the customer. Any computer machine must adapt and track patterns like punched cards overcome by solid-state drives (Baškarada et al., 2018).

The sophistication of digital applications needs both software production and efficiency upgrades. This implies, however, that the ERP architecture has discovered unavoidable faults. Which, with time, would lead to new architectures including such microservices, as opposed to itself. Any applications use this kind of construction more efficiently.

Since their development in the early '90s, legacy ERP systems are widely used. Initially developed to handle hard data, i.e., stored on hard drives or memory storage devices, ERP systems have since evolved (disparately) to address data generated on Web and IOT devices (Boniecki & Rawłuszko, 2018). One issue is that the fundamental technology that drives legacy systems is old-fashioned, unable to leverage open source software and APIs to empower interconnection. Also, they are not appropriate to an organization expanding through mergers and acquisitions. These systems cannot handle the innumerable global regulatory necessities (Brogi et al., 2018).

Consequently, the legacy systems are unable to connect easily and converse with other systems. This leads to the creation of multiple bolts-on solutions and costly both in terms of time and money (Cho & Kim, 2014). In turn, a monster legacy ERP system depends on resources with specific legacy system programming and system knowledge. These resources can be costly both in terms of time and money.

A microservice is a distinct, autonomous element contributing to a specific service. In a medium to large enterprise, many such services may combine to achieve an end goal, for example, data storage and archival. Furthermore, though robust implementation of different microservices components may improve overall efficiency, this implementation will differ based on organizational management and past knowledge (Yousif, 2016). These factors' role is not well known and needs to be studied to optimize distributed services' efficiency, i.e., microservices.

Microservices fulfills typical data storage characteristics by providing independent, expandable, and upgradeable factors fit for the evolutionary design approach. For enterprises that have traditionally used legacy ERP systems, migration to microservices will require a change in organizational thinking (Oberle & Dreiss, 2018). The distributed nature of microservices means that data structure handling and archival will be different for each service. This will require modified requirements and collaboration between teams handling each service.

Microservices are a subset of distributed computing services that offer more secure, efficient, and cost-effective alternatives to monolithic ERP systems for data archival and storage solutions in medium and large enterprises (Maas et al., 2014). As an organization scales, it will generate more data that needs to be methodically managed and supervised to be applied properly. In the medium to large enterprises, data flows in many forms and shapes.

The architectural style of Microservices has received considerable attention in recent years. The demand for micro-services began in 2014 and has continuously increased since then.

An architectural microservices solution is to create a single program as a series of small services that work with lightweight mechanisms, often an HTTP resource API. They are designed on business skills and are completely autonomous deployment machines, and can individually be used (Олещенко & Глінський, 2017). This modern architecture allows massive, complex, and scalable systems to be created, including tiny, autonomous, and highly unconnected processes that communicate with each other using APIs.

Microservice architecture.

The aim for microservices is to use autonomous units which, through decentralized container technology, such as Docker, are insulated and coordinated into a decentralized network. In normal words, this architecture paradigm's implementation often means adopting agile methodologies, such as DevOps, which decreases the time to modify the structure and extends this to the development environment.

Services are the key blocks and means of modularizing micro-service structures as the expression 'microservices' implies. Services may be deployed independently, replaced, and removed in different process circumstances (Celozzi, 2020). Any microservices focus on a single business purpose following the concept of single responsibility (SRP).

Centralized administration is discarded as far as possible and practicable, as well as data storage. This helps the team to choose the best resources, such as suitable programming languages or repositories for this mission (Molchanov & Zhmaiev, 2018). Also, without impacting other teams, decisions may be reversed or overturned.

The teams will develop new implementations of their service on their behalf through high implementation and infrastructure optimization. Implementations of microservices are stateless, except for short-term caches, which boost efficiency and durability. Sometimes, databases, in particular, are even typically run (Venugopal, 2017).

Microservices' functionality is supplemented with API calls that offer data in a format easily useable by data visualization clients. This approach lessens the code difficulties on the client-side dealing with data aggregation and transformation for visualization (Neubert et al., 2019). Therefore, Microservices can be easily implemented in small to medium organizations. However, implementing microservices in large companies needs Re-strategizing the application deployment. It can lead to self-service delivery implications since microservices need the administrators to understand the relationship between these services' demand.

Conversely, this creates the need for additional training and understanding of evolving technologies to promote their adoption and create a seamless transition. There is still limited knowledge to determine the objective and subjective factors required for adopting microservices as an alternative to legacy ERP systems.

There is a ubiquitous need in the IT sector to create quick-run quality applications. Each company thinks about the word "microservices" from large-scale cloud services to opposing start-ups (Sultan, 2020). We want to step away slowly from the conventional ERP software to loosely connected providers. Microservice architectures quickly became a must for business data management. By splitting a large collection of functions into distinct functions that allow developers to build loose-connected autonomous services as their server or utility, they are not attached to a specific server or circumstance (Plutora, 2019).

Microservices decentralize data collection choices, as well as the decentralization of decisions on logical models. Although ERP applications use a single, logical database for persistent data, companies often use a single database over various applications based on seller's licensing models.

To overcome the challenges of a monolith application development, microservices were being invented. Given its importance when deciding to take microservices as an assessment focuses for microservices adoption, the participants are asked to evaluate commonly assigned resources to microservices. One of the most key characteristics of microservices is strong scalability and stability (Laigner et al., 2021). The organization around the company plays a very significant role in adding value for the highly requested resources of Microservices architecture. Since its inception, Microservices Architectures has revolutionized the computing industry by providing optimal solutions for many unexpected complexities.

This can alone be expanded rather than the whole machine deployed. We may also reduce the database as demand falls. This prevents excessive operating costs and server failure due to high demand. The rise and decrease in operation instances may even be automated, allowing a carefree program servicing strategy to be adopted. Furthermore, microservices architecture can help to extract fast and instantaneous solutions for the current applications effectively. The microservices organizations are closely coupled, making each of them unrelated.

The use of microservices would certainly change an organization's technological and organizational culture. No modifications are therefore required to implement or alter some function in the whole codebase (Yi et al., 2019). Each provider is a specific entity without having to scale the whole application individually.

No afterthought has been given to the fact that companies with microservices have derived many benefits from this. On the other hand, turning the coin shows plainly that not all businesses are capable of the rewards of design for microservices. Be sure the company is ready to handle it before switching to microservices. Both staff and developers' resistance to microservices can be very obstacles to microservices acceptance, as microservices differ very much from the use of developers and operators (Mateus-Coelho et al., 2021). In reality, developers and operators alike will not be able to resist transition and to use microservices.

The high level of self-nomination by the team is highly responsible. The teams now will have to work with certain transversal problems traditionally handled by specialist teams. So, I think it is the right thing. Microservices are not easy to test. Each operation is directly or gradually reliant on others. Dependencies are increasing with the inclusion of more new functions.

Continuous implementation and gradual growth models enable teams to provide support quickly with microservices. Also, it can be instant when it relates to the use of utilities.

Monolithic Architecture.

Monolithic applications for various similar activities are planned. These applications are usually complex and have many strictly interconnected features. A typical method of designing applications is called monolithic architecture. A unified, indivisible entity is a monolithic program. Typically a customized user experience, a server-side program, and a database are part of this approach. It is centralized, and it operates and serves all roles at a single location.

Normally, a massive codebase and modularity are lacking in monolithic programs. To upgrade or modify things, developers accessing the same code basis. Therefore, they adjust the whole sequence at once. Monolithic implementations have strong interdependence of modules, closely interconnected (Villamizar et al., 2015). The various modules use features so that even an individual module default causes dropping dominoes to collapse, leading to the multiplicative effect.

ERP has streamlined enterprise systems and monolithic platforms share a shared data method and model covering all operations. The business requirements concentrate mainly on four fields: IT cost reduction, the productivity of enterprise applications, business procedures, and business productivity (Mosleh et al., 2018). Choose every type of ERP that is exclusive to any business. In the cloud, we built a platform that can allow one to find out which solution is the right one for the business enterprise based on comprehensive expertise in the active deployment and support of ERP applications.

Drivers for Monolithic adoption.

Overlapping problems include the issues affecting the whole program, such as recording, handling, caching, and tracking output. This category of complexity is only one feature for a monolithic framework and is thus easier to manage. Unlike the design of microservices, it is much simpler to configure and validate monolithic systems. Because a single unit is a monolithic application, end-to-end testing can be performed much quicker. The simplification of monolithic applications also makes them easy to deploy. One does not have to tackle multiple implementations when applied to monolithic systems – only one file or registry. As a common method to design software, a monolithic solution provides a team of engineers with the correct experience and skills to create a monolithic framework.

Barriers for Monolithic adoption.

It gets too difficult to grasp if a monolithic program increases. Also, it is difficult to handle a dynamic code structure within a framework. Changes in a wide and complicated program with very close connections are harder to execute. Every modification of code impacts the whole network and must therefore be coordinated extensively (Marcinauskas, 2021). This lengthens the whole construction process. A monolithic application involves the application of modern technology, which then requires rewriting the whole application.

Monolithic Vs. Microservices Systems.

The concept 'monolithic software' describes software implementation that cannot be implemented independently from the modules, as seen in Figure 2—the monolithic architecture instance. In software systems for a long time, the monolithic architecture style was the norm. Even then, some general problems with the monolithic architecture lead to conversion to microservices (Dragoni et al., 2017). The below is the list of issues:

1. Monolithic applications appear to be continuously expanding. This also increases the ambiguity, which makes it progressively difficult to retain monolithic applications. It takes a long time to identify mistakes and build new functionality (Dragoni et al., 2017).

2. If a portion of a single application is modified at some stage, it is essential to reload the entire application. This is good for smaller applications, but this may be a substantial downtime for application areas (Dragoni et al., 2017).

3. Another flaw in monolithic implementations is the management of scalability. Typically, the approach is to build more instances of the app for handling elevated load while an application is having a hit with inbound applications. Monolithic applications function as a unified, monolithic structure and cannot be divided. The application pieces, which do not require the added burden, still get it and drain money.

4. Monolithic implementations are harder to implement as there could be different standards for certain application areas than others. This means some bits are heavy in computer terms, some are heavy in memory. Developers need to select a single-size environment to fulfill both costly and suboptimal specifications (Al-Debagy & Martinek, 2019).

Microservices are unified and autonomous processes the, as described, communicate with each other to shape a spread application. Example of the architecture of microservices. They are tiny, autonomous operating systems, databases, and other supporting applications which have their remote environment. Microservices are essentially all components in an MSA program. A microservices is, for example, a webshop with microservices to treat the consumer data. It just adds, removes, updates, and lists customer details for the online store that the service does. No more roles are available to the microservices, and little else is known (Laigner et al., 2021).

It focuses solely on the small role of managing information to customers. Together, the addition of several 20 microservices is a shared framework. Microservices also interact with each other by transferring messages. This ensures that microservices can be designed according to the specifications of various programming languages and contexts.

When a corporation is set up, its implementations usually start being monolithic, depending on context. It is fair since these systems initially perform better and need less equipment under minimal circumstances. Nevertheless, they will need their technology infrastructure as businesses develop and change (Slamaa et al., 2021). When networks are expanding and dynamic, businesses become a long-term technology option for Microservices.

In this case, it is necessary to evaluate both architectures' success as an alternative to justifying such migration. The amount of memory in the operation of a procedure is memory utilization. Network output is data transfer calculation, both when transmitting and receiving the data. Wix.com has embraced microservices as part of its migration inspiration to address major technological difficulties that have caused uncertainty (Jiang et al., 2014). In 2010, the corporation began to split components into smaller services to help handle scalability. Likewise, Best Buy's architecture has been a deployment constraint. It is time to hold company online was just too long. A few decades back, companies needed to run all the server-side software such as data server administration, customized applications, network switches, and data center racks. However, with the launch of cloud computing, things got easier.

As established organizations adjust the team responsibilities to new software development practices, including the ownership of various aspects of the development cycle (Marquez et al., 2021). In the Agile Manifesto, the well-known word "self-organizing teams" will describe how many software cultures adapt them — organically and easily, with limited confusion. However, the necessary modifications could require some encouragement from leadership for other organizations. It all depends on the culture of the company experiencing the change. The layout of input as a direct result of team structure is a good way to view micro services' creation (Bucchiarone et al., 2018). The aim is to improve team frameworks, in this case, microservices, to produce products that are focused. And just about every company would take this path to Maximize the Value of the software. It works like here.

Independent Components.

Above all, all systems should be deployed and individually modified to provide more stability. Secondly, a malfunction in a single microservices only affects a certain service and does not affect the whole framework. Adding additional functionality to the microservices platform is often much faster than an ERP (Gao et al., 2020).

The architecture of Microservices offers greater mobility and facilitates the swiveling of domain areas. DevOps will concentrate on upgrading only the appropriate parts of an application by breaking up functionality to the lowest level and then resuming the associated services. The frustrating integration mechanism usually linked to ERP applications is removed. The growth of microservices is accelerated and can be done in the week and not months. Systems are typically configured to run on multiple servers (Kazanavičius & Mažeika, 2019).

Microservices operate properly with agility in all the features and functions. It means that the whole machine never falls in companies developing information infrastructure. Microservices include agility. ERP schemes have inconsistent effects on agility, and minimal effects are achieved after deployment (Tapia et al., 2020). In the past, business information planning programs helped to simplify, standardize, integrate and automate operations, thus having an unclear impact on the company's capacity to make agility.

The most simple to execute ERP architecture. The outcome is probably a monolith if no construction is implemented. ERP architecture will take an application very far as it is simple to create and helps teams bring their products before their clients very easily (Montesi et al., 2021). It has many benefits to maintain the entire codebase in one location and to launch a single program. You only have to keep one repository and can browse and find all features in one folder quickly.

The ERP design lets you deploy the approach once and only based on the existing modifications. However, the entire project will melt down if anything goes wrong.

Deployment is a dynamic method in microservices in the context of microservices architecture. The independent implementation of each micro-service is needed, extending the implementation process (Mazlami et al., 2017). Just one microservices is affected if anything goes wrong, and this will be easier to repair.

An IT team is involved in multiple platforms such as Pascal,.NET, Java, or DB2 is needed in ERP architecture maintenance. It takes much time in the monolith to find bugs and to make adjustments. Testing itself, though, is straightforward and can take place at once.

Maintenance is simpler than monolithic microservices. Smaller services also save programmers time and are easy to test. With time, productivity rises, and money saves.

If durability is involved, the monolith has little chance towards microservices. If in ERP architecture anything unexpected happens, it will interrupt the whole structure. In the meantime, splitting one service would not create major issues with the application system in the micro-service design (AL-Mandi & AL-Sharjabi, 2020).

Micro networks, however, are stable and secure in large part. Breaking one portion affects this aspect only, while the others stay unchanged. This versatility makes a high growth rate without competing with others and implementing improvements in one feature.

Due to structure complexity and size, scalability is challenging in ERP architecture to accomplish. It is difficult to update this option. Scalability is much simpler for microservices since we can only measure certain bits that need more energy. The microservices solution also benefits from the fact that each item can be individually sized. So, because the whole application must be scaled even though it does not have to be used, the entire solution is more economical and time-efficient than monoliths. Furthermore, each monolith has scalability constraints, such that the greater the number of users you buy, the more issues the monolith has (Di Francesco et al., 2019).

Many firms, however, eventually restore their ERP architectures. Contrast the simple scalability in Microservices with ERPs; when scaling is not trivial, whether the module has a sluggish internal code cannot work quicker. To scale an ERP system, a clone of the whole system must be executed on a separate computer, not removing the bottleneck of a sluggish inner stage within the monolith.

It takes a little more than microservices for ERP architecture to evolve. This is because both departments have to operate in tandem with the same code. Microservices provide quick implementation (Escobar et al., 2016). As they do in ERP architecture, teams do not have to operate parallel as any application can be supplied separately.

A single-piece arrangement is a monolith that can be divided into smaller pieces. That is why before publication, that everything should be ready. Possible issues would hamper the whole project in teamwork. Due to the microservices' structure, new capabilities can be released more rapidly by microservices (Baresi & Garriga, 2019).

Microservices are delightful in simplifying the complicated issues of attempting to modify massive, unmanageable ERPIT structures based on a vast variety of parts, technology, and applications. Monolith architecture is cheaper and quicker to build, but each particular case has to be addressed. Monoliths are an important investment for companies and are a greater challenge and a larger budget burden (Villamizar et al., 2016).

Microservices are often more costly, and the entire implementation takes longer than in monolithic applications. And they will also cost fewer, in the long term, if we consider that the working time for developers is less than a monolithic architecture.

It is equally necessary to have the ability to handle knowledge effectively in today's business environment so that it remains productive for business. Data is regarded as a highly useful competitive advantage that provides the enterprise with economic benefits. This view on data storage has been further emphasized in the progress of software development in organizations. Our research is focused on understanding the factors that promote the growth of a creative company microservice ecosystem and their contribution to organizational competitiveness.

This research helps to clarify the design strategy for improving assertiveness by considering the impact of the organizational memory components. There is a fundamental shift in how knowledge is created, used, and handled in the organizations today. It is probably obvious at this stage that the universe powered by our data would not shrink. In reality, information and data storage capacity will probably continue to increase. A dynamic phase beyond the processing and storing of information is the cognitive mechanisms of organizational memories that accumulate, perceive, and preserve information. To quickly view and summarize the results as useable knowledge at the time of a decision, organizations will have to use complex storage and recuperation procedures (Bhandary & Maslach, 2018).

Organizational remembrance is the information that has been acquired from prior experiences that may be used for decision-making. This essay discusses some of the subtleties of the memories of institutions and their impact on organizations. One key problem relevant to this thesis, which examines the facets of organizational storage, is the domain of data storage. This is currently considered an essential factor to enhance and enhance business productivity through knowledge and memory management.

The organizational memory concerns the organization's ability to take advantage of its previous events to function successfully in the present. Thus, the OM philosophy focuses on the storage and recovery processes, such that organizational and human understanding can be reused. This expertise can be stored in different deposits and is essential for enhancing the efficiency of the organization. In order to improve productivity, organizational memory enables the organization (Kaufmann et al., 2018). Its key principles are based on features to save, restore and use past business interactions. In other terms, OM learns about the background and tends to make new experiences.

Baresi, L., & Garriga, M. (2019). Microservices: The Evolution and Extinction of Web Services? Microservices, 3–28. https://doi.org/10.1007/978-3-030-31646-4_1

Baškarada, S., Nguyen, V., & Koronios, A. (2018). Architecting Microservices: Practical Opportunities and Challenges. Journal of Computer Information Systems, 1–9. https://doi.org/10.1080/08874417.2018.1520056

Berman, E. (2017). An Exploratory Sequential Mixed Methods Approach to Understanding Researchers' Data Management Practices at UVM: Findings from the Quantitative Phase. Journal of EScience Librarianship, 6(1), e1098. https://doi.org/10.7191/jeslib.2017.1098

Brogi, A., Neri, D., & Soldani, J. (2018). A microservice-based architecture for (customizable) analyses of Docker images. Software: Practice and Experience, 48(8), 1461–1474. https://doi.org/10.1002/spe.2583

Celozzi, C. (2020, December 2). How Door Dash transitioned from a code monolith to microservices. Door Dash Engineering Blog. https://doordash.engineering/2020/12/02/how-doordash-transitioned-from-a-monolith-to-microservices/

Di Francesco, P., Lago, P., & Malavolta, I. (2019). Architecting with microservices: A systematic mapping study. Journal of Systems and Software, 150, 77–97. https://doi.org/10.1016/j.jss.2019.01.001

Habadi, A., Samih, Y., Almehdar, K., & Aljedani, E. (2017). An Introduction to ERP Systems: Architecture, Implementation, and Impacts. International Journal of Computer Applications, 167(9), 1–4. https://doi.org/10.5120/ijca2017914322

Kazanavičius, J., & Mažeika, D. (2019, April 1). I am migrating Legacy Software to Microservices Architecture. IEEE Xplore. https://doi.org/10.1109/eStream.2019.8732170

Khazaei, H., Barna, C., Beigi-Mohammadi, N., & Litoiu, M. (2016). Efficiency Analysis of Provisioning Microservices. 2016 IEEE International Conference on Cloud Computing Technology and Science (CloudCom). https://doi.org/10.1109/cloudcom.2016.0051

Laigner, R., Zhou, Y., Salles, M. A. V., Liu, Y., & Kalinowski, M. (2021). Data Management in Microservices: State of the Practice, Challenges, and Research Directions. ArXiv: 2103.00170 [Cs]. https://arxiv.org/abs/2103.00170

Nawaz, N., & Channakeshavalu. (2013). The Impact of Enterprise Resource Planning (ERP) Systems Implementation on Business Performance. SSRN Electronic Journal. https://doi.org/10.2139/ssrn.3525298

Plutora. (2019, June 28). Understanding Microservices and Their Impact on Companies. Plutora. https://www.plutora.com/blog/understanding-microservices

Sampaio, A. R., Rubin, J., Beschastnikh, I., & Rosa, N. S. (2019). Improving microservice-based applications with runtime placement adaptation. Journal of Internet Services and Applications, 10(1). https://doi.org/10.1186/s13174-019-0104-0

Sandoe, K., & Olfman, L. (1992). Anticipating the mnemonic shift: Organizational remembering and forgetting in 2001. INTERNATIONAL CONFERENCE on INFORMATION SYSTEMS (ICIS), 1–12. https://core.ac.uk/download/pdf/301364184.pdf

Singh, V., & K Peddoju, S. (2017). Container-based microservice architecture for cloud applications. International Conference on Computing, Communication, and Automation (ICCCA), 847–852. https://doi.org/10.1109/CCAA.2017.8229914.

Siong Choy, C., & Yong Suk, C. (2005). Critical Factors In The Successful Implementation Of Knowledge Management. Journal of Knowledge Management Practice, 6(1), 234–258. http://www.tlainc.com/articl90.htm

Stubbs, J., Moreira, W., & Dooley, R. (2015, June 1). Distributed Systems of Microservices Using Docker and Serfnode. IEEE Xplore; 7th International Workshop on Science Gateways, Budapest, Hungary. https://doi.org/10.1109/IWSG.2015.16

J. Stubbs, W. Moreira and R. Dooley, "Distributed Systems of Microservices Using Docker and Serfnode," 2015 7th International Workshop on Science Gateways, Budapest, Hungary, 2015, pp. 34-39, doi: 10.1109/IWSG.2015.16.

Swoyer, M. L., Steve. (2020, July 15). Microservices Adoption in 2020. O'Reilly Media. https://www.oreilly.com/radar/microservices-adoption-in-2020/

Tapia, F., Mora, M. Á., Fuertes, W., Aules, H., Flores, E., & Toulkeridis, T. (2020). From Monolithic Systems to Microservices: A Comparative Study of Performance. Applied Sciences, 10(17), 5797. https://doi.org/10.3390/app10175797

Villamizar, M., Garces, O., Ochoa, L., Castro, H., Salamanca, L., Verano, M., Casallas, R., Gil, S., Valencia, C., Zambrano, A., & Lang, M. (2016). Infrastructure Cost Comparison of Running Web Applications in the Cloud Using AWS Lambda and Monolithic and Microservice Architectures. 2016 16th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid). https://doi.org/10.1109/ccgrid.2016.37