Improving IT ops efficiency is a business need

December 14, 2013

As Facebook has demonstrated, there is business value to improving efficiency of IT ops. Large organizations tend to relegate internal operational efficie to a lower priority and need to open their eyes to opportunities that open up when internal efficiencies are accomplished.

Check out Facebook IT ops

Monitoring Hadoop MapReduce Applications

April 29, 2013

While users have access to many tools that assist in performing large scale data analysis tasks, understanding the performance characteristics of their parallel computations, such as MapReduce jobs, remains difficult.  Step #1 is to create a test suite that you can reliably run after every change.

Scaling Web Applications with NoSQL

March 16, 2013

The need to scale web applications in internet speed requires the auto-sharding and “no-schema” flexibility of NoSQL databases.

Good article from Couchbase describing need for NoSQL Databases  –

Anti-Fragile DevOps

January 20, 2013

With need for rapid changes in our apps, making change hard since change is risky — is flawed.  Massive changes with rigorous change management approvals might exhibit no visible risks but could eventually lead to massive blowups catching everyone off guards.   Introducing massive changes introduce new operations vulnerabilities with each release, due to dependence upon manual process steps.

The solution is to make changes small in scope, stress the deployment process so that the release process becomes a boring and a low risk task with a heavy reliance on automation. It requires organizational acceptance to rollback the changes if an issue arises, but since the scope of change is small it becomes more manageable. This approach over time makes the product more robust & less susceptible to massive blowups.



Do we really need to focus on SLA’s?

December 12, 2012

We need to recognize that the best practice for cloud application design and configuration is to build resiliency into the application rather than expect it from the cloud platform. This way you can achieve any service-level agreement regardless of the base SLA provided by the cloud platform. Getting the performance you need is an application-specific goal anyway. What’s the value of having your sourcing and vendor management team negotiate a high and tight SLA from the cloud vendor when source of issue is the application?

First things first to extract value from Big Data

September 26, 2012

To avoid frustration, ensure there is sufficient focus on getting your big data infrastructure to perform optimally. The value of Big Data analysis will follow

Can Hadoop run on VMs ?

September 24, 2012

See below a good perspective different from VMware marketing material.



solrmeter – Stress test tool for Apache Solr – Google Project Hosting

September 23, 2012

Stress test too specifically for solr

Guide: Solr performance tuning

September 23, 2012

Article talks about hardware and software tuning. Good start but additional focus on improved performance by sharding and tuning of shards would be helpful

SPM – Scalable Performance Monitoring

September 23, 2012

Another good tool