addressalign-toparrow-leftarrow-rightbackbellblockcalendarcameraccwcheckchevron-downchevron-leftchevron-rightchevron-small-downchevron-small-leftchevron-small-rightchevron-small-upchevron-upcircle-with-checkcircle-with-crosscircle-with-pluscontroller-playcrossdots-three-verticaleditemptyheartexporteye-with-lineeyefacebookfolderfullheartglobegmailgooglegroupshelp-with-circleimageimagesinstagramFill 1light-bulblinklocation-pinm-swarmSearchmailmessagesminusmoremuplabelShape 3 + Rectangle 1ShapeoutlookpersonJoin Group on CardStartprice-ribbonprintShapeShapeShapeShapeImported LayersImported LayersImported Layersshieldstartickettrashtriangle-downtriangle-uptwitteruserwarningyahoo
This Meetup is cancelled

The Quantcast File System (QFS) -- an alternative to HDFS

  • Samsung R&D

    95 W. Plumeria, San Jose, CA (map)

    37.391071 -121.931908

  • Dear BDG members,

    This meetup we have Silvius Rus from Quantcast will be talking about Quantcast File System (QFS).  QFS is used at Quantcast in production systems at Petabyte scale.

    6pm : networking / food
    630pm - 8pm : QFS by Silvius Rus

    This event is sponsored by Samsung.
    This is a FREE event.

    This talk will describe the Quantcast File System (QFS), an alternative toHadoop¹s HDFS developed at Quantcast from the Kosmos File System (KFS) andreleased to open source in 2012.  Its main advantage is Reed-Solomonerasure coding, which saves 50% in storage costs and also pays performancedividends.  We¹ll describe the QFS architecture with emphasis onReed-Solomon, discuss implementation and operations details, and presentperformance comparisons relative to HDFS.

    Silvius Rus leads Big Data Platforms at Quantcast.  He directs, managesand participates in the development of an Internet-scale data processingsoftware stack.  Quantcast processes up to 30 petabytes of data per dayand responds in realtime to 850,000 requests per second at peak.  BeforeQuantcast he was at Google working on Gmail global load balancing,parallel memory allocation performance, server performance and C++compiler and library optimization.

Join or login to comment.


Our Sponsors

Sign up

Meetup members, Log in

By clicking "Sign up" or "Sign up using Facebook", you confirm that you accept our Terms of Service & Privacy Policy