Jumanji: The Case for Dynamic NUCA in the Datacenter
- 1 October 2020
- conference paper
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE) in 2020 53rd Annual IEEE/ACM International Symposium on Microarchitecture (MICRO)
Abstract
The datacenter introduces new challenges for computer systems around tail latency and security. This paper argues that dynamic NUCA techniques are a better solution to these challenges than prior cache designs. We show that dynamic NUCA designs can meet tail-latency deadlines with much less cache space than prior work, and that they also provide a natural defense against cache attacks. Unfortunately, prior dynamic NUCAs have missed these opportunities because they focus exclusively on reducing data movement.We present Jumanji, a dynamic NUCA technique designed for tail latency and security. We show that prior last-level cache designs are vulnerable to new attacks and offer imperfect performance isolation. Jumanji solves these problems while significantly improving performance of co-running batch applications. Moreover, Jumanji only requires lightweight hardware and a few simple changes to system software, similar to prior D-NUCAs. At 20 cores, Jumanji improves batch weighted speedup by 14% on average, vs. just 2% for a non-NUCA design with weaker security, and is within 2% of an idealized design.Keywords
This publication has 77 references indexed in Scilit:
- HeraclesPublished by Association for Computing Machinery (ACM) ,2015
- Last-Level Cache Side-Channel Attacks are PracticalPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2015
- Bubble-fluxPublished by Association for Computing Machinery (ACM) ,2013
- ZSimPublished by Association for Computing Machinery (ACM) ,2013
- The tail at scaleCommunications of the ACM, 2013
- Exascale Computing Technology ChallengesLecture Notes in Computer Science, 2011
- FlexDCPACM SIGOPS Operating Systems Review, 2009
- Cache Attacks and Countermeasures: The Case of AESLecture Notes in Computer Science, 2006
- Practical PACE for embedded systemsPublished by Association for Computing Machinery (ACM) ,2004
- An adaptive, non-uniform cache structure for wire-delay dominated on-chip cachesPublished by Association for Computing Machinery (ACM) ,2002