At the Apache Big Data conference at the Corinthia hotel in Budapest, Anjul Bhambhri, Vice President of IBM Big Data is giving her keynote on stage and is pretty excited about Spark, the high level Apache project that runs programs faster than Hadoop in memory and on disk.
“It’s the most active, top-level Apache project. We at IBM are really excited about this revolutionary technology to such an extent that we think of this as a pre-analytics operating system. Never before have so many capabilities come together on a single platform.”
In fact, in Bhambhri’s own words, the application is literally “out of this world” and is being used by NASA on space telescopes to analyse the data and look for patterns and anomalies so that “Spark can tell you whether the beeping you hear is coming from a Martian or a Mars rover”.
Bhambhri has been in the database industry for about 25 years and prior to becoming VP of IBM Big Data, she held management positions at Informix and Sybase. I decide to catch up with her later to learn more about Spark and why IBM is throwing its full support behind it. We grab one of the sofas near the coffee machines.
“The basic principle of Hadoop was to divide and conquer. It’s a massively parallel infrastructure and built on commodity hardware. So you have more data and nodes,” Bhambhri explains. “The beauty of Spark is that it’s a single platform that has brought many capabilities together and it is speeding up all applications whether the data has to be accessed from disk or memory. Which means that for customers those queries can be run much faster on terabytes or petabytes of data.”
Bhambhri is full of praise for Apache and says that part of reason Spark has been so successful is because it has a “very healthy community behind it” so it is one of those technologies that will stand the test of time.
What about some of the criticism Apache gets about the way some projects are handled?
“You don’t want bad code getting in otherwise everyone suffers. So the process of reviewing the code and having multiple people checking that it’s all done properly is pretty impressive. IBM has been a big participant in many of the Apache projects as well as open source in general. So we have it in our DNA to work in the open source community!” Bhambhri laughs.
Bhambhri is soft-spoken and friendly. She is modest when I refer to the “Tribute to Women in Technology” award she received in 2009, saying she devotes whatever time she can to encourage more women into engineering. Still, she thinks a lot more work needs to be done to increase diversity in the workforce to “tap into all the brain power” out there.
“Over the years there is definitely an upward trend but it is still very small. You have to build up that interest [in engineering] while they are still entering high school. But if they like it they should not give it up just because they don’t see enough role models.”
Bhambhri tells me that IBM has a “bring your girls to work” scheme so they can see other women in technology. Bhambhri’s daughter is studying engineering and she says that she took her to a Grace Hopper conference one year which was a real “eye-opener” for her daughter. She says these are the types of things girls should be exposed to from a young age.
Has Bhambhri faced any challenges as a women in the workforce over her long career?
Bhambhri thinks for a moment and shakes her head. “At IBM I have started many initiatives that I believed in and a lot of the people that provided support were men,” Bhambhri laughs. “I think it’s important people follow their passion because if you have passion then you have conviction.”
For now Bhambhri sees great potential for Spark in practically every industry sector, including healthcare. Through genome sequencing she thinks “every person on the planet can get personalised medicine”.
“The goal here is to work in the community and to contribute to the open source and really make this technology available to everyone so they can solve big data problems they could not solve before.”
She is particularly proud of the work IBM did a few years ago with the University of Ontario in Toronto, Canada with InfoSphere Streams technology to detect infections in premature babies.
“[We were able to] detect patterns in the data and predict the onset of an infection in a baby 24 hours in advance. That story is so near and dear to me because when the doctors get a 24 hour runway they were able to treat the babies.”
Finally, IBM’s Watson has been busy writing recipes and has even released a cookbook.
Has Bhambhri tried Watson’s recipes?
Bhambhri laughs and says she’s personally not tried them. “But I know people that have. Maybe when I get some time I will do the same!”
PREVIOUS ARTICLE«C-suite career advice: Vera Loftis, Bluewolf
Jon Collins’ in-depth look at tech and society
Phil Muncaster reports on China and beyond
Kathryn Cave looks at the big trends in tech