Sponsored by

in cooperation with

Sponsored by

Help Promote

Past Events

How to become
a SYSTOR supporter

Connect with us


How to Connect to VIRTUAL SYSTOR 2021

SYSTOR 2021 will be a virtual event.

For each day of the conference, there is a Zoom webinar and a live stream to Youtube. We recommend joining via Zoom and using Youtube as a backup in case our Zoom webinar has reached capacity.

Please note that the Poster Session on Day 1 is going to take place in a separate Zoom meeting, in which each poster has a dedicated room. Participants will join the Zoom meeting for the posters session (once the main sessions of Day 1 have concluded), and will be able to move freely between posters.

Questions to the speakers

Please post your questions to the speakers on Slack. There is a separate channel per session for this purpose. You can ask your questions live and are equally welcome to post comments and questions ahead of time.

You can join our Slack channel here: Join ACM SYSTOR Slack.

Day 1

Jun 14, 2021 17:00 Jerusalem

Join us on Zoom or on Youtube.

Webinar ID: 967 7428 3680, Passcode: 091800, International numbers available: https://acm-org.zoom.us/u/ac4S83iBiH.

Day 1 Posters Session

Jun 14, 2021 20:10 Jerusalem

Join us on Zoom.

Meeting ID: 954 1533 6866, Passcode: 220638, Find your local number: Find your local number: https://acm-org.zoom.us/u/aenyXaXR8H.

Note, the meeting room for the Posters session is separate from the webinar for the main sessions of Day 1.

Day 2

Jun 15, 2021 17:00 Jerusalem

Join us on Zoom or on Youtube.

Webinar ID: 995 8736 3277, Passcode: 486550, International numbers available: https://acm-org.zoom.us/u/aciBRajtRL.

Day 3

Jun 16, 2021 17:00 Jerusalem

Join us on Zoom or on Youtube.

Webinar ID: 960 9887 9547, Passcode: 462399, International numbers available: https://acm-org.zoom.us/u/ad6IVQuSre.

Sessions (Talks and Videos)

Day 1 Sessions

Session 1: Memory
Session Chair: Sudarsun Kannan (Rutgers University)

To Move or Not to Move? - Page Migration for Irregular Applications in Over-subscribed GPU Memory Systems with DynaMap

Lossless Instruction-to-Object Memory Tracing in the Linux Kernel

Session 2: Key-Value Stores
Session Chair: Arpan Gujarati (University of British Columbia, Vancouver

KVRAID: High Performance, Write Efficient, Update Friendly Erasure Coding Scheme for KV-SSDs

Fine-grained Control of Concurrency within KV-SSDs

KVCG: A Heterogeneous Key-Value Store for Skewed Workloads

Day 2 Sessions

Session 3: NVM Part 1
Session Chair: Sanidhya Kashyap (EPFL)

A Study of Persistent Memory Bugs in Linux Kernel

UniHeap: Managing Persistent Objects Across Managed Runtimes for Non-Volatile Memory

Session 4: Storage
Session Chair: Huaicheng Li (CMU)

BDUS: Implementing Block Devices in User Space

IMG-DNA: Approximate DNA Storage for Images

The Case for Adding Privacy-Related Offloading to Smart Storage

Reducing write amplification in flash by death-time prediction of logical block addresses

Dissecting Self-Describing Data Formats to Enable Advanced Querying of File Metadata (short paper)

Session 5: Security
Session Chair: Aastha Mehta (University of British Columbia, Vancouver)

Dealing with (Some of) the Fallout from Meltdown

S2Dedup: SGX-enabled Secure Deduplication

Length Preserving Compression – Marrying Encryption with Compression

Day 3 Sessions

Session 6: NVM Part 2
Session Chair: Aishwarya Ganesan (VMware Research)

HeuristicDB: A Hybrid Storage Database System Using a Non-Volatile Memory Block Device

SIMD-MIMD Cocktail in a Hybrid Memory Glass: Shaken, not Stirred

Session 7: Highlights
Session Chair: Ram Alagappan (VMware Research)

EvenDB: Optimizing Key-Value Storage for Spatial Locality</i>

MAD-HTLC: Because HTLC is Crazy-Cheap to Attack

Characterizing, Exploiting, and Detecting DMA Code Injection Vulnerabilities in the Presence of an IOMMU

PINT: Probabilistic In-band Network Telemetry

Self-adjusting Advertisement of Cache Indicators with Bandwidth Constraints

Session 8: Potpourri
Session Chair: Prateek Sharma (Indiana University)

JumpGate: Network Connected Acceleration for Data Analytics

LIRS2: An Improved LIRS Block Replacement Algorithm