Serverless vs. PaaS and Docker: Three-legged Hairless Yaks

Which technology is right for you?


  1. Tomi Engdahl says:

    Micro Services will replace it-lumps

    Publish often, publish quickly. You may have heard of this software development lempihokeman, even if you are a programmer. In many cases, the reality is The harsh and less flexible. Can publish only rarely and slowly.

    However, the problem is a new solution, which is surrounded by frantic hype. Micro Services realized the ideal of repeated and rapid publication. In addition, they provide many other benefits to software development as the company’s business. Developers are excited.

    “Micro-serve solution consists of independent components or services. The opposite is monolithic, application solution, which has been one of the clumps, “Solita’s senior software engineer Arto Santala says.

    “Large information systems swell like bread dough over the years. As the size increases, updating and maintenance of complicated and developers must cognitive overload, ”

    Developers want to knock down a set of monoliths inside out – as a weapon of micro-services. Their key strengths are in addition to the rapid publication of failover and technological freedom.

    Micro Services usually associated with Web services that are used by a browser in its user interface. Their implementation is utilized almost exclusively cloud services.

    Micro services are pioneers in cloud computing giant Amazon,

    “When the team is small enough understanding of the individual services and its role is all clear and it is easy to develop,” says Rautonen.

    He defines micro-services information system architectures with a single information system will be built publish the following service components. They are talking to each other a light interface, usually http

    “The most important advantages of Micro services is that the complexity of the system becomes more easily manageable and structured,”

    Fault tolerance is an extreme example of micro-services pioneer Netflix. Video service is a huge number of users, and services that require plenty of performance. If the service would be one-day disturbance, it would be a serious risk to the business. Users will not long tolerate the service, if desired video does not work at all.

    Netflix failover guarantee amounting to hundreds of micro-rise facilities, and automated rapid response to unexpected changes. Faults hidden from the user.

    Other well-known micro-services users are also music service Spotify, for example, and the world’s most popular part of the cloud-based e-commerce platform Shopify.

    If the software is not interesting, micro services you should be knowledgeable about the money. The monolith is a fast-moving market hopelessly slow. The competition stacks up to now require micro-services, which guarantee the ability to change and allow rattling in economic benefits.

    “You can lose money if you are not able to respond to changes.”

    Micro Services will also help to protect against a single supplier being trapped. The reason for this is the principle of substitution: Services will be held in such small that they can be replaced easily.

    Loose vs. tight bond

    When the service is split into independent pieces, each piece can develop independently, and not a single new version of the publication mess up anything else. The bond between the fire is the most lax. This allows for rapid publication of the model.

    Monolithic solution, the bonds are tight, everything is interconnected. If a problem or a bottleneck occurs, at least one of a change in the code lead to unpredictable risks. The change is always the version of the update, which must be tested carefully. In addition to the change in the object must be encoded into new much more.

    Micro Services resemble in many ways the concept of SOA, Service Oriented Architecture, or service-oriented architecture, which was supposed to speed up the development and integration of systems.

    “Micro Services Architecture is like the coming of a new service architecture, but there are a few big difference,”

    Soa-world architecture was the goal area, but now the architecture is not the end point, but it is a constant adjustments.

    “Micro Services Architecture SOA is like a re-awakened, but picking up only the best bits”

    “The biggest difference is that SOA was seen as bus based solution. There was talk that there are fools services, the wise bus. Micro Services will turn the other way, the intelligence services is in itself ”

    Micro Services breakthrough right now is not a coincidence.

    “Devops-cultural and automation are essential enabling factors,”

    Devops means that developers and production work together from the beginning

    One crucial change is the great popularity soared container technology, which is like the micro-services. Containers means that one and the same virtual machine can run in parallel on a number of different services that are autonomous and independent of each other. Containerization is not a mandatory condition for the micro services.

    “For example, the Amazon they are driven developers to write applications into smaller independent units of code,”

    Many developers are wary of the monolith, and go strictly the principle that from the beginning made micro-services.

    “Migrations can not be done if there is not designed from the beginning that the software runs on a micro-service”

    How micro is micro?

    Lot’s of debate also raises the concept of micro-service prefix. Many experts consider the whole concept a bad name.

    Recently, there has been talk that the micro-service must be “right-sized”
    Right vary in size from one service to another according to its own logic.

    Micro adds to the complexity

    Micro Services drift into the complexity of the swamp, where chopped too much – and the complexity of the just seek to get rid of.

    The essence is to keep the complexity of managing a highly efficient automation and monitoring for both virtual machines and containers within the configuration. Without monitoring, for example, is difficult to figure out what is due, if the interface slows down.

    “It is amateurish to make micro services without wondering services monitoring,”

    Red Hat is again ready to OpenShift cloud platform that is specifically designed for micro and services driving continuous publication. It includes a centralized view, a comprehensive automation and monitoring tools. Similar substrates are, for example, open source and IBM CloudFoundry made it onto Bluemix and Pivotal Web Services.

    Freedom and responsibility

    The new architecture provides the freedom to develop each of the micro-service tools, which are very relevant to the needs of the most effective. Team’s own preferences determine the technology of choice.

    “The only agreement are interfaces outside.”

    The commonly agreed interfaces are vital adhesive micro-services.

    Old-service to micro-cleavage services is difficult, but not impossible.

    For example, in the banking and insurance sector is still used by many decades old systems made in the COBOL programming language. New services again made the acquisition-date technology.

    “Micro Services say that it is ok. Let all the flowers bloom, when you only have interfaces. It is certainly painful to do interfacing legacy systems, but the moment cobolit are not going to leave. Flexibility they may apply for an additional layer, ”

    strict principle of micro-services is the fact that each micro service is part of their independent character of its own internal data model. Micro service need not comply with the company-wide data model.

    Relation of micro-data services to data differs from the traditional data management. Individual micro service may need to store the rest of the data in its own database format, which is reasonable for its own use. Data may temporarily be out of date and inconsistent – misunderstandings arising from the difference between the potential source of conflict between the company’s information management and software developers.


  2. Tomi Engdahl says:

    Neglected Step Child: Security in DevOps

    The use of microservices and containers like Docker have led to a revolution in DevOps. Providing the agility that business have long awaited, these new technologies also introduce inherent security implications that cannot be ignored at a time when the enterprise attack surface continues to grow wider. Let’s consider these risks and how organizations can minimize their exposure to them.

    According to a recent report by 451 Research, nearly 45% of enterprises have either already implemented or plan to roll out microservices architectures or container-based applications over the next 12 months. This confirms the hype surrounding these emerging technologies, which are meant to simplify the life of application developers and DevOps teams. Microservices can break down larger applications into smaller, distinct services; whereby containers in this context are viewed as a natural compute platform for microservices architectures.

    Microservices and containers enable faster application delivery and improved IT efficiency. However, the adoption of these technologies has outpaced security. A recent research study by Gartner (DevSecOps: How to Seamlessly Integrate Security into DevOps) shows that fewer than 20% of enterprise security teams have engaged with their DevOps groups to actively and systematically incorporate information security into their DevOps initiatives.

  3. Tomi Engdahl says:

    Serverless computing could unleash a new startup ecosystem

    While serverless computing isn’t new, it has reached an interesting place in its development. As developers begin to see the value of serverless architecture, a whole new startup ecosystem could begin to develop around it.

    Serverless isn’t exactly serverless at all, but it does enable a developer to set event triggers and leave the infrastructure requirements completely to the cloud provider. The vendor delivers exactly the right amount of compute, storage and memory and the developer doesn’t even have to think about it (or code for it).

    In the beginning we had physical servers, but there was lots of wasted capacity. That led to the development of virtual machines

    Then came containers, which really began to take off with the development of Docker and Kubernetes

    More recently, we’ve seen the rise of serverless or event-driven computing. In this case, the whole idea of infrastructure itself is being abstracted away.

    While it’s not truly serverless, since you need underlying compute, storage and memory to run a program, it is removing the need for developers to worry about servers.

    All of the major vendors have launched serverless products with AWS Lambda, Google Cloud Functions and Microsoft Azure Functions all offering a similar approach.


Leave a Comment

Your email address will not be published. Required fields are marked *