Wednesday, October 24, 2012

Cloud Computing definition

This time I'm gonna once more talk about Cloud and scalability. There are couple of misunderstandings in this topic, so I'll try to explain it.

Developing scalable applications is based on several techniques, but all of this techniques are connected in some way. They're all using compute power of distributed computers. One of the most popular solution is Cloud Computing which is a step forward from Fabric, Grid and Cluster Computing which are using distributed machines, but there are not scalable capabilites in them.

This new technique haven't growed up in one day. It's all about evolution. It all started from SOA (Service Oriented Arhitecture) - which is model of organizing software into loosely coupled separate components which are talking together through contracts. Many companies started to rebuild their solutions to work in that way. It's easier to build and maintain systems that have been built from few of smaller blocks. Another step of evolution were fast Clusters which are offering big amount of compute resources. But quickly it has turned out that these clusters are generating really big costs, so companies that owned them tried to found solution for this. This solution is today called Cloud Computing - which is renting compute and storage resources to other companies or even individuals.

Every successfull application sooner or later will have to face the "Internet scale". It is very important to prepare services for this effect. Applications must be scalable enough to serve millions of users in the same time. Developing this kind of applications isn't trivial, and there are many things that must have been taken into account by architects during planning architecture of whole system.

There are several models of cloud services and many vendors. It's hard to decide which company or which model would be best for current requirements. Cloud isn't solution for everything - there are some types of applications which would benefit from cloud solutions:
  • applications with big nonstructural data
  • streaming media content
  • SOA components
Autoscaling these types of applications can be very easy in the cloud. But first of all we need to define what the real Cloud is. There are many definitions of it but the most popular is the definition from NIST organization. Here are some key characteristics of Cloud Computing (cited from enclosured link):
  • On-demand self-service. A consumer can unilaterally provision computing capabilities, such as server time and network storage, as needed automatically without requiring human interaction with each service provider.
  • Broad network access. Capabilities are available over the network and accessed through standard mechanisms that promote use by heterogeneous thin or thick client platforms (e.g., mobile phones, tablets, laptops, and workstations).
  • Resource pooling. The provider’s computing resources are pooled to serve multiple consumers using a multi-tenant model, with different physical and virtual resources dynamically assigned and reassigned according to consumer demand. There is a sense of location independence in that the customer generally has no control or knowledge over the exact location of the provided resources but may be able to specify location at a higher level of abstraction (e.g., country, state, or datacenter). Examples of resources include storage, processing, memory, and network bandwidth.
  • Rapid elasticity. Capabilities can be elastically provisioned and released, in some cases automatically, to scale rapidly outward and inward commensurate with demand. To the consumer, the capabilities available for provisioning often appear to be unlimited and can be appropriated in any quantity at any time.
  • Measured service. Cloud systems automatically control and optimize resource use by leveraging a metering capability at some level of abstraction appropriate to the type of service (e.g., storage, processing, bandwidth, and active user accounts). Resource usage can be monitored, controlled, and reported, providing transparency for both the provider and consumer of the utilized service.
So only when we have all of this mentioned above elements we can name our solution as a Cloud.

Tuesday, October 23, 2012

ScrumBut

There are a lot materials about agile software development. Tons of books how quick software development can be. But how it really looks inside? How does it look from the developers perspective?

In most cases it looks as like that:


Agile teams are mostly working with Scrum, because it's really easy to implement it, and it's really easy to work with. But to be honest neither do I or any of my developer fellows know a single company that is really implementing it fully. Of course everybody has a daily meetings, but sometimes there are exceptions to this rule, everybody knows that we need to do retrospective, but not every time, etc. It's really a fact, that most companies have customized version of Scrum. There's even a name for it: ScrumBut (http://www.scrum.org/ScrumBut). It's really funny that laziness of some managers and developers has gained its own name.

Scrum isn't the only methodology that is being badly customized. It is a really common way of doing software. I think that there are some explanations for this strange fact. Firstly we do have two worlds meeting here - managers and IT. Developers have their own language and they really don't see the need of managing them - people don't like to be controlled and monitored all the time (daily meetings may have this discomfort of impossibility to hide that we've done nothing productive today) and from the other perspective we have some strange dressed guys that have been doing something all day long, and we even can't tell what was it exactly...

Finding a solution for this problem isn't very easy. Mostly there are communication problems - managers have zero knowledge about IT, and developers have zero knowledge about project management. Both sides are very different and both sides need to cooperate together. It would be great when both sides can think about the other camp and how to make life of both sides easier.