All posts by Navin Kabra

Pune Municipal Corporation gets CIO, new website, wiki

PMC has gone hi-tech. Yesterday, they unveiled a new, improved website that is expected to be more userfriendly. They can now apply for birth and death certificates online and pay municipal taxes online. PMC is the first corporation to have a payment gateway (through ICICI bank and HDFC bank) according to CIO Anupam Saraph.

Yes, Pune has a CIO – Chief Information Officer. The PMC collaborated with SEAP, the Software Exporters Association of Pune, to appoint Anupam, who will guide the direction of all hi-tech goodness that is to happen in PMC soon. Pune is also the first municipal corporation to have a CIO.

“The website has been a voluntary effort by the software companies in Pune such as Persistent software, Eclipse software who provided manpower purely on a voluntary basis,” said Saraph.

The other important feature in the website is the inclusion of Wiki software that allows creation of an employee zone for internal management of the corporation.

Through Wiki, employees would now be able to create, edit, link, and organize the content of the projects they are involved in.

To ensure better navigation, the PMC website will change every month. “In the next version Wiki software will allow the citizens to talk about development plans of the city. Later versions intend to include the software for citizens to help the corporation with their strategic infrastructure expertise, so as to develop a standard in the city,” said Saraph.

“Till now we have trained 180-odd employees and expect to train 50 employees a month to ensure more usage of the software,” he said.

The website in it present form has been visited by more than half a million people.

Source: ZDNet India

News in Brief: Harbinger makes insect games, Zensar gets new business, Aryabhatta award for Dr. Kale

Harbinger to design and develop games to teach children about insects

Pune-based Harbinger Knowledge Products has partnered with a US non-profit organization, the Entomological Foundation, to design an develop games based online learning software products targeted towards 3rd and 4th grade children in the US. The games will teach the children about “the exciting world of insects and their role in our environment.”

Harbinger are also the creators of Flockpod, a web-service that allows web publishers to create a space on their webpage where users can interact with each other right on-the-spot, without leaving the page, and of Raptivity an library of pre-built flash interactive videos that can be used for e-Learning.

Source: press release

Zensar wins $5 million in new contracts

Zensar Technologies, software services and BPO provider, has won over $5 million new deals in West Asia and South Africa in the first three months of 2008 and is now targeting over ten per cent of its total revenues from this market segment.

Sify business

Aryabhatta Award for Pune’s Dr. Kale

The Astronautical Society of India (ASI), Bangalore, has on Wednesday announced its annual awards for the year 2006 in which Dr. Pramod Kale, Director, Integrated Circuit and Information Technology Pvt. Ltd., Pune, has been chosen for the prestigious Aryabhata Award.

The Aryabhatta award was instituted to recognise talented Indian individuals who have made significant contributions to astronautics. The press release doesn’t actually list Dr. Kale’s contributions which won him this award.

Source: Newstrack India

Events: Pune OpenCoffee Club meeting & Startup “lunch” combined today

The Pune OpenCoffee Club and the Pune Startup “Lunch” events are being combined into one event to be held today (Friday, 4th April) at 5:30pm on the terraces of Bookeazy on BMCC road. (No, there will be no lunch at 5:30pm.)

The Pune OpenCoffee Club (POCC) is an informal club for people interested in the Pune startup ecosystem to meet regularly for discussions, exchange of ideas and networking. The Pune Startup Lunch is gathering where people interested in working for startups can meet founders of startups to get an idea of the possibilities.

Mukesh Singhal of Canaan Partners will also attend to mingle with entrepreneurs. He is also interested in meeting with startups on Saturday. Send him an e-mail at msinghal@canaan.com if you are interested.

The event is free and open for all. You should attend if you are:

  • A founder of a startup
  • Interested in starting a startup someday
  • Interested in working for a startup
  • Generally interested in the Pune startups ecosystem

Details of venue and directions are here. See here to get an idea of who have already confirmed attendance attending.

Pubmatic and Komli to power all eBay.in ads

eBay India), today announced an exclusive advertising relationship with Indian ad network Komli Media. Komli is now the exclusive seller of all of eBay India’s banner advertising inventory up to end March 2009.

In addition, eBay India will also use Komli’s Pubmatic ad optimization service to improve ad revenues. Pubmatic can estimate the revenues that a website can earn from any particular ad, and thus improve revenues by comparing ads from different ad networks (e.g. Google Adsense, Yaho! Publisher Network, or even Komli itself) and then only showing those ads that are likely to generate the most revenue. This service (which is available free of cost to any web publisher) will now get used on eBay India’s pages.

The Pubmatic service is developed in Komli’s engineering center in Pune.

Company Profile: Amberpoint

Company profile of AmberPoint from the Punetech wiki

AmberPoint is a provider of middleware for governing the runtime of Service Oriented Architecture (SOA) applications. Utilizing a policy-based approach to managing the health and well-being of SOA-based applications, AmberPoint’s software offers comprehensive capabilities for visibility into and control of loosely coupled systems.

Contents

[edit] Main Features

[edit] Application Discovery

Automatically figure out what all SOA applications exist in an environment and determine their capabilities. Upload this information to a central repository. Keep the repository updated as applications come and go and their capabilities change. Track dependencies between various applications.

Knowing what applications are deployed, and the dependencies between them is the first step towards managing the applications properly.

[edit] Policies

Once applications have been discovered and cataloged AmberPoint allows administrators to specify policies which need to be enforced against the applications. And enforce them. Also, architects can define approved policies and make them available to their SOA developers. Developers can select the policies they need to implement and attach them to their services on the fly, without any additional coding effort.

[edit] Monitoring

Reports: Monitor the various components of the SOA in real time. Produce reports about things like: throughput, availability, response times, and errors/faults. This is achieved without requiring any modifications to existing services (no code changes), or any extra development time. And with low overhead.

Composite service monitoring: In addition to monitoring a single component, you also want to be able to monitor the same parameters for a composite service created using these components.

SLAs: The ultimate aim of this monitoring is to ensure that Service Level Agreements are being met.

Visualization: Show pretty pictures to get an idea of who is using what. And how much. And the dependencies between various components. And the flow of data.

Early warnings: Based on information about the existing SLAs and how the applications are composed of sub-components, it is possible to figure out which SLAs are likely to get violated based on the predicted trends for some of the low-level measurements taken at one or more of the components. This can help in catching problems before they become problems. Also, using the same basic information, traffic spikes can be prevented by throttling can be applied at the appropriate places at the apropriate times.

[edit] Exception Handling and Root Cause Analysis

Provide an integrated view of message flows that occur across the various components of the system. Thus allow a single interface where administrators can search for errors/issues by message type, content, client credentials, and even tie the low level messages to higher-level business transactions. This eliminates the need to manually piece together a message flow by searching multiple log files.

Also, allow for rule-based routing of exceptions to various exception handlers. It integrates with existing exception-handling resources such as in-house exception-management frameworks as well as exception-handling workflows and process-management systems.

[edit] Testing

When one component changes, it is possible that the composite service might break. And testing for this is difficult because of the distributedness and loose-coupling that is inherent in a service oriented architecture. Using AmberPoint, administrators can automatically verify the performance and functionality of a changed service against applications that consume it and the services on which it is dependent. Additionally, they can perform “what-if” runs to see the potential impact of management policies (especially security) as they change within the runtime environment. AmberPoint also allows simulation of services for testing purposes using actual production traffic and realistic performance characteristics.

[edit] Regulatory Compliance

AmberPoint provides pieces needed for complying with regulations in the financial industry, health industry, credit card industry and the SOX regulations. Examples include multi-factor authentication, role based access control, etc..

Website: http://www.amberpoint.com/

(Editor’s note: Occassionally I will be posting such quick’n’dirty profiles of Pune-based companies. These will essentially be information collected from the company website. This is not intended to be an in-depth look at the technology. It is more directed towards people how might not know about the existence of the company, or not know what it does.)

Pune OpenCoffee Club Meeting – 4th April (for Startups)

If you are a Pune-based entrepreneur (or think you’d like to be an entrepreneur someday), consider joining the Pune OpenCoffee Club. There is a meeting of the POCC on Friday. Main agenda is for the members to get to know each other. Also a couple of VCs will be attending – they also want to get to know Pune entrepreneurs.

Details from Anjali Gupta (anjali at NOSPAM bookeazy dot com):

Dear Pune OpenCoffee members,

Thank you for your enthusiastic response in bootstrapping the Club. In the last few weeks over 50 members and 20 startups have joined POCC on http://punestartups.ning.com

Who’s here?
Pringoo, Komli, Mithi, Mangospring, Zmanda, 42, ThinkingSpace, TechZemplary, Coriolis, Josh Software, Niyuj, Indyvation, Mobikon, 5 Colors, Business Gateway, Spring Computing, TastyKhana, Lipikaar, and BookEazy.

We’ve planned our first networking meetup on 4th April, 5:30pm – 7.30pm (Friday) at BookEazy.

We’ve also invited a few guests who have expressed interest in meeting POCC entrepreneurs at the meetup.
– Mukul Singhal, Canaan Partners
– Niren Shah, Norwest Venture Partners
– Startup mentors and incubators will also be attending

Do invite entrepreneurs, founders, investors/advisers or those interested in starting up or joining startups.

For event details and directions to the venue, please visit:
look here

Don’t forget to confirm your participation here to help us plan:
RSVP link: http://upcoming.yahoo.com/event/428897/
RSVP over email: anjali at NOSPAM bookeazy dot com

Feel free to add your feedback and comments on the POCC forums.

Look forward to meeting everyone.

Best regards,
Anjali

Pune needs more of a tech “community”. So please join.

Building EKA – The world’s fastest privately funded supercomputer

Eka, built by CRL, Pune is the world’s 4th fastest supercomputer, and the fastest one that didn’t use government funding. This is the same supercomputer referenced in Yahoo!’s recent announcement about cloud computing research at the Hadoop Summit. This article describes some of the technical details of Eka’s design and implementation. It is based on a presentation by the Eka architects conducted by CSI Pune and MCCIA Pune.

Interconnect architecture

The most important decision in building a massively parallel supercomputer is the design of how the different nodes (i.e. processors) of the system are connected together. If all nodes are connected to each other, parallel applications scale really well (linear speedup), because communication between nodes is direct and has no bottlenecks. But unfortunately, building larger and larger such systems (i.e. ones with more and more nodes) becomes increasingly difficult and expensive because the complexity of the interconnect increases as n2. To avoid this, supercomputers have typically used sparse interconnect topologies like Star, Ring, Torus (e.g. IBM’s Blue Gene/L), or hypercube (Cray). These are more scalable as far as building the interconnect for really large numbers of nodes is concerned. However, the downside is that nodes are not directly connected to each other and messages have to go through multiple hops before reaching the destination. Here, unless the applications are designed very carefully to reduce message exchanges between different nodes (especially those that are not directly connected to each other), the interconnect becomes a bottleneck for application scaling.

In contrast to those systems, Eka uses an interconnect designed using concepts from projective geometry. The details of the interconnect are beyond the scope of this article. (Translation: I did not understand the really complex mathematics that goes on in those papers. Suffice it to say that before they are done, fairly obscure branches of mathematics get involved. However, one of these days, I am hoping to write a fun little article on how a cute little mathematical concept called Perfect Difference Sets (first described in 1938) plays an important role in designing supercomputer interconnects over 50 years later. Motivated readers are encouraged to try and see the connection.)

To simplify – Eka uses an interconnect based on Projective Geometry concepts. This interconnect gives linear speedup for applications but the complexity of building the interconnect increases only near-linearly.

The upshot of this is that to achieve a given application speed (i.e. number of Teraflops), Eka ends up using fewer nodes than its compatriots. This means it that it costs less and uses less power, both of which are major problems that need to be tackled in designing a supercomputer.

Handling Failures

A computer that includes 1000s of processors, 1000s of disks, and 1000s of network elements soon finds itself on the wrong side of the law of probability as far as failures are concerned. If one component of a system has a MTBF (mean time between failures) of 10000 hours, and the system has 3000 components, then you can start expecting things to fail once every 10 hours. (I know that the math in that sentence is probably not accurate, but ease of understanding trumps accuracy in most cases.)

If an application is running on 500 nodes, and has been running for the last 20 hours, and one of the nodes fails, the entire application has to be restarted from scratch. And this happens often, especially before an important deadline.

A simple solution is to save the state of the entire application every 15 minutes. This is called checkpointing. When there is a failure, the system is restarted from the last checkpoint and hence ends up losing only 15 minutes of work. While this works well enough, it can get prohibitively expensive. If you spend 5 minutes out of every 15 minutes in checkpointing your application, then you’ve effectively reduced the capacity of your supercomputer by 33%. (Another way of saying the same thing is that you’ve increased your budget by 50%.)

The projective geometry architecture also allows for a way to partition the compute nodes in such a way that checkpointing and status saving can be done only for a subset of the nodes involved. The whole system need not be reset in case of a failure – only the related subset. In fact, with the projective geometry architecture, this can be done in a provably optimal manner. This results in improved efficiency. Checkpoints are much cheaper/faster, and hence can be taken more frequently. This means that the system can handle failures much better.

Again, I don’t understand the details of how projective geometry helps in this – if someone can explain that easily in a paragraph or two, please drop me a note.

The infrastructure

The actual supercomputer was built in just 6 weeks. However, other aspects took much longer. It took an year of convincing to get the project funded. And another year to build the physical building and the rest of the infrastructure. Eka uses

  • 2.5MW of electricity
  • 400ton cooling capacity
  • 10km of electrical cabling
  • 10km of ethernet cabling
  • 15km of infiniband cabling

The computing infrastructure itself consists of:

  • 1800 blades, 4 cores each. 3Ghz for each core.
  • HP SFS clusters
  • 28TB memory
  • 80TB storage. Simple SATA disks. 5.2Gbps throughput.
  • Lustre distributed file-system
  • 20Gbps infiniband DDR. Eka was on the cutting edge of Infiniband technology. They sourced their infiniband hardware from an Israeli company and were amongst the first users of their releases – including beta, and even alpha quality stuff.
  • Multiple Gigabit ethernets
  • Linux is the underlying OS. Any Linux will work – RedHat, SuSe, your favorite distribution.

Its the software, stupid!

One of the principles of the Eka project is to be the one-stop shop for tackling problems that require huge amounts of computational powers. Their tagline for this project has been: from atoms to applications. They want to ensure that the project takes care of everything for their target users, from the hardware all the way up to the application. This meant that they had to work on:

  • High speed low latency interconnect research
  • System architecture research
  • System software research – compilers etc.
  • Mathematical library development
  • Large scientific problem solving.
  • Application porting, optimization and development.

Each of the bullet items above is a non-trivial bit of work. Take for example “Mathematical library development.” Since they came up with a novel architecture for the interconnect for Eka, all parallel algorithms that run on Eka also have to be adapted to work well with the architecture. To get the maximum performance out of your supercomputer, you have to rewrite all your algorithms to take advantages of the strengths of your interconnect design while avoiding the weaknesses. Requiring users to understand and code for such things has always been the bane of supercomputing research. Instead, the Eka team has gone about providing mathematical libraries of the important functions that are needed by applications specifically tailored to the Eka architecture. This means that people who have existing applications can run them on Eka without major modifications.

Applications

Of the top 10 supercomputers in the world, Eka is the only system that was fully privately funded. All other systems used government money, so all of them are for captive use. This means that Eka is the only system in the top 10 that is available for commercial use without strings attached.

There are various traditional applications of HPC (high-performance computing) (which is what Eka is mainly targeted towards):

  • Aerodynamics (Aircraft design). Crash testing (Automobile design)
  • Biology – drug design, genomics
  • Environment – global climate, ground water
  • Applied physics – radiation transport, supernovae, simulate exploding galaxies.
  • Lasers and Energy – combustion, ICF
  • Neurobiology – simulating the brain

But as businesses go global and start dealing with huge quantities of data, it is believed that Eka-like capabilities will soon be needed to tackle these business needs:

  • Integrated worldwide supply chain management
  • Large scale data mining – business intelligence
  • Various recognition problems – speech recognition, machine vision
  • Video surveillance, e-mail scanning
  • Digital media creation – rendering; cartoons, animation

But that is not the only output the Tatas expect from their investment (of $30 million). They are also hoping to tap the expertise gained during this process for consulting and services:

  • Consultancy: Need & Gap Analysis and Proposal Creation
  • Technology: Architecture & Design & Planning of high performance systems
  • Execution: Implement, Test and Commissioning of high performance system
  • Post sales: HPC managed services, Operations Optimization, Migration Services
  • Storage: Large scale data management (including archives, backups and tapes), Security and Business Continuity
  • Visualization: Scalable visualization of large amounts of data

and more…

This article is based on a presentation given by Dr. Sunil Sherlekar, Dr. Rajendra Lagu, and N. Seetha Rama Krishna, of CRL, Pune, who built Eka. For details of their background, see here. However, note that I’ve filled in gaps in my notes with my own conjectures, so errors in the article, if any, should be attributed to me.

News in Brief: VMWare, QLogic to invest; Symantec interviews own employees; Aar-em lights up villages

VMWare to invest $100 Million in India

VMware Inc. announced today that it plans to invest more than $100 million in India by 2010, partly to double its research and development workforce in that country to more than 1,000 employees. VMWare currently has R&D centers in Bangalore and Pune. While a lion’s share of the new investment will go to Bangalore, some of it will also be used to expand VMWare’s Pune facility.

Source: VMWare Press Release

QLogic announces Development Lab in Pune

QLogic Corp. (Nasdaq:QLGC), a leader in networking for storage and high performance computing (HPC), today announced the opening of the QLogic International Development Lab (QIDL) in the heart of Pune, India. QLogic is hiring engineers now for the QIDL, located in the ICC Trade Towers off Senapati Bapat Road. According to their press release the QIDL development team is expected to help “design the future” of networking for storage and HPC. For details, contact hr-india@qlogic.com.

Source: TechNewsRelease.

Symantec Pune conducts interviews of own employees

It is common practice to conduct exit interviews of employees leaving a company to find out why they are leaving and thus identifying weaknesses of an organization. Symantec, Pune is turning the concept on its head by conducting interviews of employees who have been with the company for a long time, and focusing on the strengths of the organization that made them stay back.

After selecting leaders from this group of Symantec loyalists, their example and achievements were put before the others on the company intranet. “This benefited us. Now these leaders have not only got recognition for their extra-efficiency, it has also encouraged others to see whether they can be incorporated in this list of leaders,” says Abhay Valsangkar, Symantec India’s head of HR.

Source: Sify Business.

Aar-em Electronics Pvt. Ltd. a Pune based firm known for its Champion brand of Uninterruptible Power Supplies now has entered the field of providing Uninterruptible light source for villages after sunset by providing them lights and other utilities powered by solar energy.

“MOHRI” a village situated at 100 kms from Pune City Maharashtra, India having 27 houses and a population of 160 of which 75 are children today boasts of having electricity which has been provided by Aar-em Electronics Pvt. Ltd. Where once life practically came to a standstill at sundown are now abuzz with activities. The solar-based devices have illuminated the villagers’ houses like never before- kerosene lamps are a thing of the past now. Even, infotainment is beamed to them through solar-powered television sets.

Aar-em is offering a system which typically contains a roof-installed solar PV module, storage battery, charge controller, interior wiring, and switches and fixtures with the capacity to power two- low-wattage high power LED modules.

Source: Business Wire India

Yahoo and CRL (EKA) to collaborate on cloud computing research

(Link courtesy Amit Paranjape via e-mail)

Yahoo and CRL have just announced that they will collaborate on research into cloud computing. I believe this announcement is essentially about Yahoo’s use of Hadoop on CRL’s EKA supercomputer.

This Yahoo!/CRL announcement comes on the eve of the first ever Hadoop Summit. Apache Hadoop is a Free Java software framework that supports data intensive distributed applications running on large clusters of commodity computers. It enables applications to easily scale out to thousands of nodes and petabytes of data. In Feb 2008, Yahoo announced that its Search Webmap application has been converted to a Hadoop application that runs on a more than 10,000 core Linux cluster and produces data that is now used in every Yahoo! Web search query.

CRL’s EKA supercomputer is a Hewlett-Packard based system with 14,400 processors, 28 terabytes of memory, 140 terabytes of disks, a peak performance of 180 trillion calculations per second (180 teraflops), and sustained computation capacity of 120 teraflops for the LINPACK benchmark. Of the top ten supercomputers in the world, EKA is the only supercomputer funded by the private sector that is available for use on commercial terms.

This announcement should also increase the interest in attending the presentation on EKA to be held here (i.e. in Pune) in a few days (Thursday, 27th March). The talk promises to be very interesting, especially given the background of speakers Dr. Lagu and Dr. Sherlekar (who used to be profs in IIT-B amongst other things). Details.