Big Data Computations: Comparing Apache HAWQ, Druid, Google Spanner and GPU Databases

Jul 2017

24 Mon

25 Tue

26 Wed

27 Thu 08:15 AM – 10:00 PM IST

28 Fri 08:15 AM – 06:25 PM IST

29 Sat

30 Sun

Make a submission

MLR Convention Centre, Whitefield, Bengaluru,

##Theme and format
The Fifth Elephant 2017 is a four-track conference on:

Data engineering – building pipelines and platforms; exposure to latest open source tools for data mining and real-time analytics.
Application of Machine Learning (ML) in diverse domains such as IOT, payments, e-commerce, education, ecology, government, agriculture, computational biology, social network analysis and emerging markets.
Hands-on tutorials on data mining tools, and ML platforms and techniques.
Off-the-record (OTR) sessions on privacy issues concerning data; building data pipelines; failure stories in ML; interesting problems to solve with data science; and other relevant topics.

The Fifth Elephant is a conference for practitioners, by practitioners.

Talk submissions are now closed.

You must submit the following details along with your proposal, or within 10 days of submission:

Draft slides, mind map or a textual description detailing the structure and content of your talk.
Link to a self-record, two-minute preview video, where you explain what your talk is about, and the key takeaways for participants. This preview video helps conference editors understand the lucidity of your thoughts and how invested you are in presenting insights beyond your use case. Please note that the preview video should be submitted irrespective of whether you have spoken at past editions of The Fifth Elephant.
If you submit a workshop proposal, you must specify the target audience for your workshop; duration; number of participants you can accommodate; pre-requisites for the workshop; link to GitHub repositories and documents showing the full workshop plan.

##About the conference
This year is the sixth edition of The Fifth Elephant. The conference is a renowned gathering of data scientists, programmers, analysts, researchers, and technologists working in the areas of data mining, analytics, machine learning and deep learning from different domains.

We invite proposals for the following sessions, with a clear focus on the big picture and insights that participants can apply in their work:

Full-length, 40-minute talks.
Crisp, 15-minute talks.
Sponsored sessions, of 15 minutes and 40 minutes duration (limited slots available; subject to editorial scrutiny and approval).
Hands-on tutorials and workshop sessions of 3-hour and 6-hour duration where participants follow instructors on their laptops.
Off-the-record (OTR) sessions of 60-90 minutes duration.

##Selection Process

Proposals will be filtered and shortlisted by an Editorial Panel.
Proposers, editors and community members must respond to comments as openly as possible so that the selection processs is transparent.
Proposers are also encouraged to vote and comment on other proposals submitted here.

Selection Process Flowchart

We will notify you if we move your proposal to the next round or reject it. A speaker is NOT confirmed for a slot unless we explicitly mention so in an email or over any other medium of communication.

Selected speakers must participate in one or two rounds of rehearsals before the conference. This is mandatory and helps you to prepare well for the conference.

There is only one speaker per session. Entry is free for selected speakers.

##Travel grants
Partial or full grants, covering travel and accomodation are made available to speakers delivering full sessions (40 minutes) and workshops. Grants are limited, and are given in the order of preference to students, women, persons of non-binary genders, and speakers from Asia and Africa.

##Commitment to Open Source
We believe in open source as the binding force of our community. If you are describing a codebase for developers to work with, we’d like for it to be available under a permissive open source licence. If your software is commercially licensed or available under a combination of commercial and restrictive open source licences (such as the various forms of the GPL), you should consider picking up a sponsorship. We recognise that there are valid reasons for commercial licensing, but ask that you support the conference in return for giving you an audience. Your session will be marked on the schedule as a “sponsored session”.

##Important Dates:

Deadline for submitting proposals: June 10
First draft of the coference schedule: June 20
Tutorial and workshop announcements: June 20
Final conference schedule: July 5
Conference dates: 27-28 July

##Contact
For more information about speaking proposals, tickets and sponsorships, contact info@hasgeek.com or call +91-7676332020.

Hosted by

The Fifth Elephant

The Fifth Elephant - known as one of the best data science and Machine Learning conference in Asia - has transitioned into a year-round forum for conversations about data and ML engineering; data science in production; data security and privacy practices. more

All submissions

Previous Next

Big Data Computations: Comparing Apache HAWQ, Druid, Google Spanner and GPU Databases

Submitted Apr 18, 2017

Section: Full talk for data engineering track Technical level: Intermediate

A class of big data computations known as the distributed merge tree was required to be built to aggregate user information across multiple data sources in media domain. This class is characterized by non-scalar aggregates all the way to the root of the merge tree – equivalent of a Set union operation in SQL at every level of the tree. Typical big data technologies were mostly supporting only scalar aggregates. The Set union operation must be implemented outside of the data store, resulting in non-standard implementation and consequent inefficiencies.

To illustrate the problem, we built a prototype on top of Druid, one of the claimants to the throne of analytical data processing. Druid supports only scalar aggregates and as a result, the Set union operation had to be implemented at the application level. Data transfer into and out of Druid as well as the complexity of thread processing at the Java layer led to inefficiencies – resulting in 200+ seconds for the computation in this case.

Apache HAWQ with its ability to perform multi-dimensional partitioning of the data, support for full SQL queries (consequently, it supports Set union operations) and its efficient distributed query optimization techniques looked like the ideal candidate for this use case. However, HAWQ’s dependence on Hadoop as the underlying file system plus the inherent complexity of the computation led to poorer than expected results – HAWQ took about 100 seconds to process the same query, but the SLA was less than 10 seconds.

However, it turned out that the multi-dimensional partitioning was inefficient – we overcame the problem by having multiple HAWQ clusters and an intelligent client that stores meta-data to route queries to appropriate clusters. By ensuring each HAWQ cluster is independent, we were able to bring down the time to execute the query to 20 seconds.
Finally, we have implemented the same query with a GPU database (Kinetica in particular) to benchmark its performance on an Amazon g2.8x instance. Finally, the response time for the same query was around 12 seconds – a bit more optimization would help meet the SLA.

We shall also be evaluating Google Spanner and comparing with all the datastores above to illustrate how efficiently Spanner supports SQL and how easy/difficult it was to solve complex big data computations.

Outline

Key audience takeaways include:
Overview of media campaign management tool in the business context and problem being solved.
Introduction to Druid, Apache HAWQ and Kinetica.
Introduction to Google Spanner and the ACID transactions it supports through TrueTime API.
Performance benchmark comparing all the four data stores on AWS cloud infrastructure.

Requirements

Nothing specific.

Speaker bio

Dr. Vijay Srinivas Agneeswaran has a Bachelor’s degree in Computer Science & Engineering from SVCE, Madras University (1998), an MS (By Research) from IIT Madras in 2001, a PhD from IIT Madras (2008) and a post-doctoral research fellowship in the LSIR Labs, Swiss Federal Institute of Technology, Lausanne (EPFL). He has joined as Director of Technology in the data sciences team of SapientNitro. He has spent the last ten years creating intellectual property and building products in the big data area in Oracle, Cognizant and Impetus. He has built PMML support into Spark/Storm and realized several machine learning algorithms such as LDA, Random Forests over Spark. He led a team that designed and implemented a big data governance product for a role-based fine-grained access control inside of Hadoop YARN. He and his team have also built the first distributed deep learning framework on Spark. He is a professional member of the ACM and the IEEE (Senior) for the last 10+ years. He has four full US patents and has published in leading journals and conferences, including IEEE transactions. His research interests include distributed systems, data sciences as well as Big-Data and other emerging technologies. He has been an invited speaker in several national and International conferences such as O’Reilly’s Strata Big-data conference series. He will also be speaking at the Strata Big-data conference in London in May 2017. He has given the keynote speech in the fifth elephant conference in 2014. He lives in Bangalore with his wife, son and daughter and enjoys researching history and philosophy of Egypt, Babylonia, Greece and India.