Nov 2024
Our HotInfra ‘24 paper demonstrates how combining diverse compute types can save costs while handling load fluctuations faced by online services.
July 2024
Our upcoming MICRO’24 paper introduces a novel branch predictor organization unlocking hierarchical branch predictor designs.
Jun 2024
Amna Shahab is now Dr. Amna Shahab. Congratulations!!
Apr 2024
We introduce Hydra at SESAME 2024, a hybrid architecture that combines VMs with serverless for microservice load resilience.
Sep 2023
Our upcoming MICRO’23 paper tackles the problem of cold microarchitectural state on a context switch.
May 2023
Antonis Katsarakis receives the prestigious EuroSys Roger Needham PhD Award - Honorable Mention.
May 2023
The SESAME workshop on serverless systems, applications and methodologies at EuroSys was a huge success. Thanks to all who participated!
Jan 2023
We introduce Saba, an application-aware bandwidth allocation scheme for datacenter networks. To appear at EuroSys’23!
Jan 2023
Jukebox has been selected as an IEEE Micro Top Picks Honorable Mention for 2023.
Oct 2022
Our HPCA’23 paper introduces a new BTB organization that, for a given storage budget, significantly improves the effective BTB capacity over prior proposals.
March 2022
Our upcoming ISCA’22 paper analyzes serverless functions on modern hardware, identifying new challenges and specialization opportunities.
Feb 2022
Join the 1st Serverless & vHive tutorial at ASPLOS’22, which brings experts from the industry & academia and includes both talks and hands-on sessions.
Sep 2021
Ever wondered what happens under the hood of Serverless clouds? Our provider-agnostic benchmarking framework, called STeLLAR, to feature in IISWC’21. Paper is available now.
Sep 2021
Congrats Dr. Artemiy Margaritov and Dr. Antonis Katsarakis on their successful PhD defense!
Jul 2021
We revisit the question of invalidating vs updating cache coherence protocols for tomorrow’s cache hierarchies. Our paper to appear in PACT’21!
Apr 2021
Our vHive/REAP work receives a Distinguished Artifact Award at ASPLOS’21 (2 out of 75 papers)!
Feb 2021
We introduce Zeus: a fault-tolerant replication protocol that exploits locality for high performance. To appear at EuroSys’21!
Jan 2021
We release vHive, a full-stack open-source framework for serverless experimentation and innovation. Check our ASPLOS’21 paper and join the vHive open-source community now!
Jan 2021
Hermes – our high-performance fault-tolerant replication protocol [ASPLOS’19] – has been selected as a Top Pick Honorable Mention!
Nov 2020
Two papers accepted to ASPLOS’21. One characterizes and optimizes snapshot-based serverless function starts using our open-source end-to-end serverless framework; the other reduces page walk latencies in virtualized cloud workloads via a simple patch to the Linux kernel.
Jul 2020
Our latest work shows how RDMA network predictability can be turned into a high-rate covert communication channel. For details, check out Bankrupt at WOOT’20, co-located with USENIX Sec’20. Proof-of-Concept on GitHub.
Apr 2020
Congrats to our lab’s first graduate Dr. Priyank Faldu!
Mar 2020
Our idea paper on speeding up training of deep neural nets with evolutionary algorithms will appear in GECCO’20!
Mar 2020
We open-sourced the Hermes protocol (ASPLOS’20). Find out more on the website.
Jan 2020
Our ISPASS’20 paper evaluates an InfiniBand switch and shows that providing both low latency and high bandwidth is a challenge.
Nov 2019
Need fast, fault-tolerant and strongly-consistent data replication in the cloud? Meet Hermes. Appearing at ASPLOS 2020.
Nov 2019
Our latest work makes the case for Domain-Specialized Cache Management for Graph Analytics. In HPCA’20.
Sep 2019
Priyank Faldu awarded the first place in the Student Research Competition at PACT 2019 for his work on a hardware/software co-design to accelerate graph analytics.
Sep 2019
Amna Shahab receives the Best Poster Award at the 2019 ARM Research Summit for her work on on using neuroevolution to speed up ML training.
Aug 2019
Our IISWC’19 paper is taking A Closer Look at Lightweight Graph Reordering.
Jul 2019
Our upcoming MICRO’19 paper introduces Prefetched Address Translation, a new approach for cutting the latency of TLB misses.
Jun 2019
Honored to be elected to the Board of Directors of ACM SIGARCH.
Feb 2019
Stretch wins the Best Paper Award at HPCA 2019!
Nov 2018
Our idea paper on accelerating virtual address translation via learned indexes will be presented at Workshop on ML for Systems at NIPS, 2018.
Nov 2018
How does one reconcile throughput and quality-of-service on a multi-threaded core running colocated server workloads? Stretch shows the way! Appearing in HPCA 2019.
Sep 2018
Amna Shahab, a 3rd-year PhD student in our lab, received the Best Poster award at this year’s ARM Research Summit!
Jul 2018
Our upcoming MICRO 2018 paper makes the case for all-private cache hierarchies for server CPUs. Say good-bye to shared last-level caches!
Jun 2018
Promoted to an Associate Professor (aka Reader in British academic parlance).
Jan 2018
Aggressive replication, strong consistency guarantees and high throughput in a scale-out cluster are not mutually exclusive. Surprised? Read our EuroSys 2018.
June 2017
Shotgun revolutionizes control flow delivery for server processors. Appearing at ASPLOS 2018.
Sep 2017
Oracle has generously supported our research on scale-up graph processing for future memory systems.
June 2017
Our work on variability-aware cache management will appear at PACT 2017.
May 2017
Leeway has been selected to compete in the Cache Replacement Championship in June.
Apr 2017
The Mondrian Data Engine will engage at ISCA 2017.
Oct 2016
Boris has been inducted into the MICRO Hall of Fame!
Oct 2016
Boomerang accepted to HPCA 2017. It shows how to design a high-performance core front-end without costly metadata - great news for server processors!
June 2016
Two papers accepted to MICRO 2016.
May 2016
Our paper debunking the usefulness of dead block prediction for last-level caches will appear in WDDD.
May 2016
Fat Caches for Scale-Out Servers has been accepted for publication in IEEE Micro.
Mar 2016
The IEEE Micro Special Issue on Near-Data Processing, which I guest co-edited, is now available.
Oct 2015
Asynchronous Memory Access Chaining will appear in VLDB 2016.
Sep 2015
Confluence: Unified Instruction Supply for Scale-Out Servers will appear in MICRO 2015.
May 2015
Sort vs. Hash Join Revisited for Near-Data Execution will appear in ASBD 2015.
Mar 2015
Manycore Network Interfaces for In-Memory Rack-Scale Computing accepted to ISCA 2015!
Dec 2014
Our Workshop on Near-Data Processing (WoNDP) was hugely successful at this year’s MICRO with over 100 people in attendance!
Sep 2014
Bulk Memory Access Prediction and Streaming accepted to MICRO 2014!
Sep 2014
The IEEE Micro Special Issue on Big Data, which I guest co-edited, is now available.
Aug 2014
Recipient of the Google Faculty Research Award.
Apr 2014
Serving on the ASPLOS, MICRO, and HPCA program committees. Please consider submitting to these top venues in computer systems research.
Dec 2013
Meet the Walkers gets Best Paper Runner-Up kudos at MICRO!