Hadoop Users Group UK

Name: September Meetup
Start: 2013-09-10T18:00:00Z
Location: King

September Meetup

Sep 10, 2013 · 7:00 PM
King

Our September meetup is kindly hosted by King.com. We'll have the usual drinks and food from 6:30pm onwards, talks will be starting around 7pm.

As this is hosted in their office we need to give names to security, so please make sure to RSVP if you are going to come.

The three talks for the evening are:

• Dr Relational or: How I Learned to Stop Worrying and Love the Database (Andy Done, Data Warehouse Lead, King)

In the face of explosive growth King's Hadoop data warehouse simply wasn't scaling fast enough. Find out why King is extending its Big Data platform with MPP database ExaSol and processing its data 100s of times faster.

• Facebook Scale Graph Processing using Giraph (Nitay Joffe, Software Engineer, Facebook)

This talk will discuss the design and architecture of Giraph - an offline distributed graph processing system built on top of Hadoop. We have scaled Giraph to process very large graphs, at Facebook we run PageRank on a 400 billion edge graph in a matter of minutes. We will dive into the architecture that allows Giraph to scale to very large data sizes.

• GraphChi (Michael Leznik, Head of BI - London, King)

GraphChi, a disk-based system for computing efﬁciently on graphs with billions of edges. By using a well-known method to break large graphs into small parts, and a novel parallel sliding windows method, GraphChi is able to execute several advanced data mining, graph mining, and machine learning algorithms on very large graphs, using just a single consumer-level computer.

Join or login to comment.

Hit enter to add your reply

Salih O.

First 2 technical oriented presentations are related with Graph processing but the last one (Andy Done) is exactly what we need. They don't remove hadoop/hive it is a hybrid use case. They still have Hive but on top of that(or side by side) they have ExaSol. In near future we cannot implement just hadoop or just DB we need to find hybrid methods to solve the complex business use cases.

Thanks King and organisers for this great event.
Salih

0 · September 11, 2013
- David A.
  
  Yes, I agree with Salih and perhaps this might also partially answer the posts of Jonathan and Mark earlier - my take was that King were highlighting a very real business use case within the Hadoop ecosystem - what to do about SQL-based analytics on structured data. Another was streaming/near real-time analytics. These are still good topics for an Hadoop User Group because you may have to consider integrations or other solutions for different use cases. King still want to land most stuff into Hadoop, as far as I could tell. They have unstructured data needs too. Some technologies may not be quite mature enough yet, so they've gone with Exasol for faster structured data analytics. There are other cluster/columnar solutions out there that can integrate with Hadoop too (I won't plug :-)) and also some interesting "new" projects e.g. Spark/Shark, StingerYARN/Tez (next gen Hive)
  
  0 · September 18, 2013
Dan H.

You can find the three slide decks here:

http://www.slideshare.net/nitayj/20130910-giraph-at-london-hadoop-users-group

http://www.slideshare.net/huguk/graphchi-big-graph-processing

http://www.slideshare.net/huguk/king-hug-uk

Enjoy!

1 · September 11, 2013
Kaho L.

Thanks for presentations organized by King. I wonder where I can download the slides.

0 · September 11, 2013
Afzal S

Same here agreed with Mark since there were audience from different background. But all in all I enjoyed it and look forward for more.

0 · September 11, 2013
Jonathan G.

I'd agree with Mark Samson below. It was a great event and many thanks to King for hosting. The technical detail was great but it would be good to understand too what the user need was that drove the technology. Some context around the applications would make some of the tech demands and constraints much more clear. Thanks again though for all the work organising & see you all again soon.

0 · September 11, 2013
Mark S.

The two graph talks were very technical and really needed to include some information on the business use cases. The talk from King was interesting but it seems strange to include a talk that is primarily about moving away from Hadoop in a Hadoop User Group. Generally it might be better to have less talks (perhaps 2 rather than 3) and more time for networking.

0 · September 11, 2013
Alex M.

Thanks for three interesting presentations - plus several good chats. And thanks to King for the venue and pizza.

0 · September 10, 2013
Sameera N.

Apologies can't make it tonight. If anyone looking for a good accountant then please get in touch with me at [masked]

0 · September 10, 2013
A former member

Looking forward to my first HUG in a while. I'm interested to talk to anyone doing near-real-time ML, involving Hadoop -- we've got a few interesting things cooking for you

1 · September 10, 2013
pravin

Apologies, I can't make it today.

0 · September 10, 2013
Catherine H.

I'm sorry I won't be able to make it anymore, the talks would have been great to see.

0 · September 10, 2013
Ben W.

Apologies, I can't make it any more. Any advice on how to beat level 78 would be greatly appreciated.

0 · September 10, 2013
Bill C.

Bill Carless
Sales Manger Broadberry Data Systems

0 · August 27, 2013
Sushil

Looking forward to it

0 · August 21, 2013
Tariq H.

How do I handle a terabyte per second?

0 · August 19, 2013
Peter H.

I'm a big data expert.

0 · August 19, 2013