BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Vienna
BEGIN:DAYLIGHT
DTSTART:20200329T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20191027T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260404T020023Z
UID:5e09f5544602c144941008@ist.ac.at
DTSTART:20200224T100000
DTEND:20200224T110000
DESCRIPTION:Speaker: Oana Balmau\nhosted by Dan Alistarh\nAbstract: Cloud s
 torage stacks are being challenged by new workloads\, new hardware and new
  performance requirements. First\, workloads evolved from following a read
 -heavy pattern (e.g.\, a static web-page) to a write-heavy profile where t
 he read:write ratio is closer to 1:1 (e.g.\, as in the Internet of Things)
 . Second\, the hardware is undergoing rapid changes.  The divide between f
 ine-grained volatile memory and slow block-level storage is rapidly being 
 bridged by the emerging byte-addressable non-volatile memory devices and t
 he fast block-addressable NVMe SSDs (e.g.\, Intel Optane NVMe SSDs). Third
 \, performance requirements in storage systems now emphasize low tail late
 ncy\, in addition to high throughput. In this talk I will argue that exist
 ing storage systems have fundamental limitations that do not allow them to
  fully meet these challenges\, and that therefore the storage stack needs 
 to undergo radical change. In particular\, using state-of-the-art key-valu
 e stores I will demonstrate that with modern workloads and hardware the bo
 ttleneck shifts from I/O to CPU\, invalidating an assumption that has unde
 rpinned all past storage system design. In line with this observation I wi
 ll then present a new design paradigm for key-value stores that departs fr
 om the conventional wisdom of optimizing disk usage and instead optimizes 
 CPU usage. To do so\, we keep data unsorted on disk\, reduce contention fo
 r shared data structures\, and do away with expensive maintenance operatio
 ns.This design has been implemented in the KVell key-value store. KVell ou
 tperforms state-of-the-art key-value stores such as RocksDB in both read- 
 and write-heavy workloads\, running on modern NVMe SSDs.  Thanks to its no
 vel design\, KVell achieves up to 5x better throughput\, and up to two ord
 ers of magnitude lower tail latency.
LOCATION:Mondi Seminar Room 2\, Central Building\, ISTA
ORGANIZER:tguggenb@ist.ac.at
SUMMARY:Oana Balmau: Redesigning Storage Systems for Future Workloads\, Har
 dware\, and Performance Requirements
URL:https://talks-calendar.ista.ac.at/events/2615
END:VEVENT
END:VCALENDAR