BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Hardware and Software Co-design for HTAP Workloads
DTSTART:20180626T143000
DTEND:20180626T163000
DTSTAMP:20260408T071359Z
UID:281017fcb02f058bd2b4b9ebd33a826e1f4c09b3383dbcf3e10187a4
CATEGORIES:Conferences - Seminars
DESCRIPTION:Zilu Tian\nEDIC candidacy exam\nExam president: Prof. Christop
 h Koch\nThesis advisor: Prof. Babak Falsafi\nCo-examiner: Prof. Paolo Ienn
 e\n\nAbstract\nReal-time analytics has gained more visibility in recent ye
 ars. Instead of relying on historical data for decision making\, some busi
 ness intelligence applications demand analysis on fresh data to gain criti
 cal insights\, highlighting the rise of mixed Online Transaction Processin
 g (OLTP) and Online Analytical Processing (OLAP) workloads\, which are coi
 ned Hybrid Transactional/Analytical Processing (HTAP)  workloads. The OLT
 P workload is characterized by short-lived transactions\, where data is st
 ored in a row-oriented database. The OLAP workload consists of long-runnin
 g\, scan-heavy\, ad-hoc queries\, where data is stored in a column-oriente
 d layout. The OLTP and OLAP workloads have conflicting needs for latency a
 nd throughput. No single physical layout fits the hybrid nature of HTAP wo
 rkloads. In this project\, we explore hardware and software co-design to a
 ccelerate HTAP workloads.\n\nBackground papers\nGather-Scatter DRAM:In-DRA
 M Address Translation to Improve the Spatial Locality of Non-unit Strided 
 Access\, by Vivek Seshadr et al.\nColumn-Stores vs Row-Stores: How Differe
 nt Are They Really? by Daniel J. Abadi et al.  \nBridging the Archipelago
  between Row-Stores and Column-Stores for Hybrid Workloads\, by Joy Arulra
 j et al.
LOCATION:BC 04 https://plan.epfl.ch/?room==BC%2004
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR