Max-Planck-Institut für Informatik
max planck institut
mpii logo Minerva of the Max Planck Society

MPI-INF or MPI-SWS or Local Campus Event Calendar

<< Previous Entry Next Entry >> New Event Entry Edit this Entry Login to DB (to update, delete)
What and Who
Title:Caribou -- Intelligent Distributed Storage for the Datacenter
Speaker:Zsolt Istvan
coming from:ETH Zurich
Speakers Bio:Zsolt Istvan is a recent PhD graduate of the Systems Group at ETH Zurich. His research looks at

using FPGAs in the context of databases and distributed systems, with the goal of building hybrid
solutions and specialized accelerators for data intensive tasks. Before graduate school he was a
Master's student at ETH Zurich, Switzerland, and a Bachelor's student at the
Technical University of Cluj-Napoca, Romania.

Event Type:SWS Colloquium
Visibility:SWS, RG1, MMCI
We use this to send out email in the morning.
Level:AG Audience
Date, Time and Location
Date:Thursday, 8 February 2018
Duration:90 Minutes
Building:E1 5
In the era of Big Data, datacenter and cloud architectures decouple compute and storage resources from each other for better scalability. While this design choice enables elastic scale-out, it also causes unnecessary data movements. One solution is to push parts of the computation down to storage where data can be filtered more efficiently. Systems that do this are already in use and rely either on regular server machines as storage nodes or on network attached storage devices. Even though the former provide complex computation and rich functionality since there are plenty of conventional cores available to run the offloaded
computation, this solution is quite inefficient because of the over-provisioning of computing capacity and the bandwidth mismatches between storage, CPU, and network.  Networked storage devices, on the
other hand, are better balanced in terms of bandwidth but at the price of offering very limited options for offloading data processing.

With Caribou, we explore an alternative design that offers rich offloading functionality while matching the available line-rate processing performance of either storage or network. It also does this in a much more efficient package (size, energy consumption) than regular servers. Our FPGA-based prototype system has been designed such that the internal data management logic can saturate the network for most operation mixes, without being over-provisioned. As a result, it can extract and process data from storage at multi-GB/s rate before sending it to the computing nodes, while at the same time offering features such as replication for fault-tolerance.

Caribou has been released as open source. Its modular design and extensible processing pipeline make it a convenient platform for exploring domain-specific processing inside storage nodes.
Name(s):Claudia Richter
EMail:--email address not disclosed on the web
Video Broadcast
Video Broadcast:YesTo Location:Kaiserslautern
To Building:G26To Room:111
Meeting ID:
Tags, Category, Keywords and additional notes
Attachments, File(s):

Claudia Richter/MPI-SWS, 02/05/2018 10:28 AM
Last modified:
Uwe Brahm/MPII/DE, 02/08/2018 07:01 AM
  • Claudia Richter, 02/05/2018 10:34 AM -- Created document.