TensorFlow: A system for large-scale machine learning

December 16, 2016November 11, 2019 ~ Adrian Colyer ~ 3 Comments

TensorFlow: A system for large-scale machine learning Abadi et al. (Google Brain) OSDI 2016 This is my last paper review for 2016! The Morning Paper will be taking a two week break for the holidays, resuming again on the 2nd January. Sometime inbetween I’ll do a short retrospective on the year. It seems fitting to ... Continue Reading

FaSST: Fast, scalable and simple distributed transactions with two-sided (RDMA) datagram RPCs

December 15, 2016November 11, 2019 ~ Adrian Colyer ~ 6 Comments

FaSST: Fast, scalable and simple distributed transactions with two-sided (RDMA) datagram rpcs Kalia et al., OSDI 2016 Back in January I wrote a short piece entitled ‘All change please’ looking at some of the hardware changes making their way to our datacenters and the implications. One of those changes is super-fast networking (as exploited by ... Continue Reading

SCONE: Secure Linux containers with Intel SGX

December 14, 2016November 11, 2019 ~ Adrian Colyer ~ 13 Comments

SCONE: Secure Linux Containers with Intel SGX Arnautov et al., OSDI 2016 We looked at Haven earlier this year, which demonstrated how Intel’s SGX could be used to shield an application from an untrusted cloud provider. Today’s paper choice, SCONE, looks at how to employ similar ideas in the context of containers. …existing container isolation ... Continue Reading

Diamond: Automating data management and storage for wide-area, reactive applications

December 13, 2016November 11, 2019 ~ Adrian Colyer ~ Leave a comment

Diamond: Automating data management and storage for wide-area, reactive applications Zhang et al., OSDI 2016 Diamond tackles the end-to-end problem of building reactive applications, defined here as those that update end-user visible state without requiring any explicit user action: … today’s popular applications are reactive: they provide users with the illusion of continuous synchronization across ... Continue Reading

XFT: Practical fault-tolerance beyond crashes

December 12, 2016November 11, 2019 ~ Adrian Colyer ~ 4 Comments

XFT: Practical fault-tolerance beyond crashes Liu et al., OSDI 2016 Here’s something that’s been bugging me for a while now. The state of the art in security has moved from the assumption of a secured perimeter and a trusted environment inside the firewall to a notion of perimeter-less security. It’s pretty much impossible to keep ... Continue Reading

DQBarge: Improving data-quality tradeoffs in large-scale internet services

December 9, 2016November 11, 2019 ~ Adrian Colyer ~ Leave a comment

DQBarge: Improving data-quality tradeoffs in large-scale Internet services Chow et al. OSDI 2106 I'm sure many of you recall the 2009 classic "The Datacenter as a Computer," which encouraged us to think of the datacenter as a warehouse-scale computer. From being glad simply to have such a computer, the bar keeps on moving. We don't ... Continue Reading

Just say NO to Paxos overhead: replacing consensus with network ordering

December 8, 2016November 11, 2019 ~ Adrian Colyer ~ 19 Comments

Just say NO to Paxos overhead: replacing consensus with network ordering Li et al., OSDI 2016 Everyone knows that consensus systems such as Paxos, Viewstamped Replication, and Raft impose high overhead and have limited throughput and scalability. Li et al. carefully examine the assumptions on which those systems are based, and finds out that within ... Continue Reading

Interlude: On assumptions

December 7, 2016November 11, 2019 ~ Adrian Colyer ~ 9 Comments

This post started out as an introduction to tomorrow's paper choice, but when it grew too long, I decided to move it to a post all of its own... Let's talk about assumptions for a moment. Often implicit (if you're really lucky, they might be made explicit and described in a 'system model' or similar), ... Continue Reading

The SNOW theorem and latency-optimal read-only transactions

December 6, 2016November 11, 2019 ~ Adrian Colyer ~ 3 Comments

The SNOW theorem and latency-optimal read-only transactions Lu et al., OSDI 2016 Consider a read-only workload (as in 100%). You can make that really fast - never any need to coordinate, never any need to invalidate any cached values… Now consider a write-only workload - you can make that even faster, if no-one’s ever going ... Continue Reading

REX: A development platform and online learning approach for runtime emergent software systems

December 5, 2016November 11, 2019 ~ Adrian Colyer ~ Leave a comment

REX: A development platform and online learning approach for runtime emergent software systems Porter et al. OSDI 2016 If you can get beyond the (for my taste, ymmv) somewhat grand claims and odd turns of phrase (e.g., “how the software ‘feels’ at a given point in time” => metrics) then there’s something quite interesting at ... Continue Reading

the morning paper

a random walk through Computer Science research, by Adrian Colyer
Made delightfully fast by strattic

Uncategorized