Sunday, May 14, 2017

5 Things I learned during the RedHat Summit


I attended Red Hat Summit earlier this month. I accidentally ran into a few of my old colleagues by accident and everyone was wondering where I have been for the last few years! Yes.. There is linked in that I occasionally visit and some local meet ups that I attend. But it is still pretty hard to keep up with. It has been a while ( 5-6 years actually) I haven't blogged about anything. So here I am blogging about some recent experiences ...

Red Hat Summit was a great community meetup with lots of interesting discussions and presentations around new (and also not-so-new) technologies. With 5000+ attendees in attendance and most of them coming from midsize and large companies, it was great to interact with a number of people. These guys were not your typical technology enthusiasts that you meet at a tech Meetup but technical leaders from established companies with ambitions to grow and adopt/make relevant technologies for growth . Here are my top 5 takeaways...

1. Enterprises are moving to the cloud at a much faster clip
It was great to see enterprises of all sizes moving to the cloud as fast as they can. These were multi-billion dollar financial institutions and government agencies talking about how they are overcoming traditional barriers like security, governance and regulatory compliance during the migration process. Keeping applications agnostic to underlying deployment infrastructure is very critical for a number of businesses. There were also presentations and discussions about using a middleware platforms like OpenStack and OpenShift container platform eliminates the direct dependency on cloud providers. Executives from cloud service providers like Amazon, Microsoft, Google, Rackspace and CenturyLink all were on the stage describing the support they are providing to make the transition successful. My favorite presentation was MS SQL Server running on Linux containers hosted on Azure and being accessed by dockerized apps running MS .NET and java apps alike.


2. Containers have already started to rule the app world

It was amazing to see a lot of momentum behind docker and containerization of the deployment platforms. Again these are not those Silicon Valley startups but intelligent banks like Deutsche bank moving to the containerization platforms. This is going to be an interesting area to follow for the next several years until it becomes the mainstream.



3. Kafka has become the defacto messaging solution
Move over MQ … companies of all sizes are looking at Kafka as the next generation alternative. With the growth in the data and messages, it is becoming more important to handle the data. Coupled with the microservices philosophy of 'dumb pipes', architects are loathe to push the orchestration and mediation to the middleware hub style solutions.

4. DevOps in NOT only about build and deployment automation
These days a number of companies are pitching software products that allow to create and manage the devops pipeline. However the pipeline is just one part of it. Ultimately it is a combination of tools, people and processes that yield true results. Educating and empowering the teams, setting up feedback loop, changing the mindset and making stakeholders comfortable in their new riles are the most challenging aspects. Often it turns out to be the real challenge where adoption fails. It was great to see several banks sharing their journey to DevOps and how they moved away from ‘create ticket’ mode to ‘self service’ as a measure of true agility.

5. Planning as-we-know-it is dead !
This comes from RedHat CEO.  It is a philosophical take and change in mindset around how companies tackle the problems and build solutions. The days of ‘traditional’ approach of 5 year strategic plans and long release cycles are over.  The companies that are successful today are the ones reinventing themselves with iterative cycle of Try -> Learn ->Improve. Actually this is practically the mode with which every startup works but it is a refreshing to see companies of all sizes buying into the same approach via thinking 'lean' and MVPs.

You can catch some of action here



Sunday, May 22, 2011

LinkedIn: Congratulations ! You are 1 in 100 million !!!

Congratulations for being one in a 100 million users. In March this year LinkedIn reached milestone of 100 million registered users and I am sure you are one of those 100 million. The press release said -
We’re growing at roughly one million new LinkedIn members every week – that’s more than one new professional joining the network every second !

Here is some more info if you are interested -


The State of LinkedIn 2011


While there is no doubt LinkedIn is hugely successful, the fact remains that it is only 1/5th or 1/6th size of that famous social network called Facebook. I know that LinkedIn is mainly targeted towards professionals but what strikes me is that amount of time a user spend on LinkedIn may not even compare with that on Facebook. In other words, LinkedIn is missing the stickiness that you would expect from these social sites that makes users visit them over and over again.

Curious to find out, what is missing I did a quick round of LinkedIn. I see that it has features like Groups, Company News and LinkedIn Answers. I would imagine they have right featureset for what they are meant for. Also it has basic app marketplace. But where are the apps that would be of interest to me? Take something like Hiring. Hiring is a very social activity since a lot of positions are filled based on networking and trust. Where is an app that for that? I do not see it. May be there is something like it available if you pay for a premium membership. But my point is that I believe apps like these would be of immense help to all the users.

One final point, now that LinkedIn had a very successful IPO. The stock doubled right after the opening. This buying frenzy ended up making the company valuation to be around $10B. If you do the calculation, that means every user is valued $100. Compare that to Microsoft's Skype acquisition which valued each user to be worth roughly around $13. I am not a stock expert nor I claim to be one. So rather than pondering over why is that the case, at least let's enjoy our moment of glory as one of the 100 million LinkedIn users !

Sunday, February 27, 2011

Will IBM Watson turn into Dr. Watson?

IBM's Watson computer put on quite a show earlier this month in the game show Jeopardy ! Well, if you have not heard about Jeopardy, there is a good chance that you are not based in North America because it is a US based show. The unique thing about this show is that it not a trivial question and answers type of quiz program but it has questions with a lot of puns. That really makes it pretty hard for somebody (like me!) to get the answers right quickly ! Watson was pitted against 2 of the all time champions in Jeopardy. So by no means it was an ordinary match up. Watson came out on top after an interesting battle. You can catch glimpse here




As you would have guessed, Watson is an artificial intelligence computer system capable of answering questions posed in natural language, developed in IBM's DeepQA project. For the technology enthusiasts, here is some information that will excite you. Watson is made up of a cluster of ninety IBM Power 750 servers with a total of 2880 POWER7 processor cores and 16 Terabytes of RAM. Watson's software runs thousands of algorithms on the questions it receives in parallel and then compares the answers to decide the best answer by attaching a percentage value. Watson's programming is written in both Java and C++ and uses Apache Hadoop distributed file system, Apache UIMA (Unstructured Information Management Architecture) framework.

We should herald Watson's feat as a new breakthrough in artificial intelligence. Watson demonstrates great ability to perform natural language processing. Watson breaks the questions into parts and tries to understand the exact question that it needs to answer. Then it uses a number of analytical techniques to come up with right answers.

While Jeopardy proved to be a great showcase for Watson to unleash its glory, it was not just built for the jeopardy showdown. After all it came with a huge price tag. Its unique ability to do natural language and analytical processing makes it a workhorse to solve array of problems. There are countless possibility in which it could be leveraged. There are some obvious areas where it could be of immense value.

IBM is reported working in healthcare with some hospitals, doctors, researchers and universities to build innovative solution in the medical field. We cannot expect Watson to diagnose the illness on its own (yet), however it can be an excellent tool that the doctors that would help them make better decisions.

For example, Watson might recognize that the kidney failure in our patient is linked to kidney failure in a patient in Buffalo and another in San Antonio; all three patients, he might inform me, were taking a “natural” weight loss supplement that contained a Chinese herb, aristolochia, that has been associated with more than 100 cases of kidney failure.

Other areas where it could be of immediate help would be customer call centers, employee help desks, equity portfolio analysis and so on. This link even claims that it would be great tool for match making !

It is too early to predict how IBM and others would be able to turn Watson into something more meaningful than a machine that vanquished its human opponents. But I sincerely think feel that it has potential to do that !