본문

서브메뉴

Reinventing Datacenter System Stacks for Resource Harvesting
Reinventing Datacenter System Stacks for Resource Harvesting
Reinventing Datacenter System Stacks for Resource Harvesting

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211152830
ISBN  
9798384082132
DDC  
004
저자명  
Qiao, Yifan.
서명/저자  
Reinventing Datacenter System Stacks for Resource Harvesting
발행사항  
[Sl] : University of California, Los Angeles, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
233 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
주기사항  
Advisor: Xu, Harry Guoqing;Kim, Miryung.
학위논문주기  
Thesis (Ph.D.)--University of California, Los Angeles, 2024.
초록/해제  
요약The rise of cloud computing and recent AI breakthroughs have radically expanded the demand for datacenter hardware resources, including CPU, memory, and accelerators such as GPUs. Despite the critical need to improve resource utilization and reduce operational cost, current datacenter system stacks-comprising OSes and runtime systems-struggle to fully utilize hardware resources due to high load variability and stringent performance requirements of datacenter workloads, leading to substantial waste of compute and memory resources.This dissertation demonstrates that it is feasible to safely and efficiently harvest stranded datacenter resources, even when they are intermittently available and dispersed across servers. Specifically, we identify two previously overlooked resource harvesting opportunities in today's data center system stacks. First, although datacenter applications often have varying and potentially large resource demands, they typically include elastic components that can be safely discarded under resource pressure, making them ideal for utilizing idle resources with temporal availability. Existing operating systems and runtime systems, though, lack proper interfaces for applications to convey such semantics and take advantage of idle resources. Second, while the availability of resources per server is unpredictable, combining stranded resources across servers can offer better overall availability. However, this opportunity is unavailable to many datacenter workloads that were designed for running on a single machine.Driven by these insights, this dissertation rethinks the datacenter system stack and introduces holistic designs for OS abstractions, the OS kernel, and application runtime systems for resource harvesting. The contributions of this dissertation are fourfold.First, we investigate how to harvest resources, especially memory which is inelastic and hard to re-assign between applications, within a single server. We introduce Midas, an OS memory abstraction that allows applications to use idle memory for storing their soft state. Midas efficiently manages soft memory with a kernel-runtime co-design, achieving near-optimal performance for four real-world datacenter applications and responding to extreme memory pressure quickly enough to avoid running out of memory.Second, we explore how to harvest resources across servers. We present Hermit, a redesigned OS kernel paging/swap system that enables applications to harvest idle memory on remote servers with full transparency and efficiency. Hermit allows any application to harness remote memory without changing a single line of code, making it practical for legacy real-world datacenter applications. It also achieves three orders of magnitude lower tail latency and up to 1.87 times higher throughput for latency-critical and batch-processing applications, respectively.Third, built atop Hermit, Canvas is a resource isolation mechanism for the kernel swap system that allows multiple applications to share remote memory without performance interference. By segregating resource usages and access patterns of co-running applications, Canvas further adaptively optimizes kernel swap for each application. Our evaluation and performance study with a wide range of datacenter applications demonstrate that Canvas reduces performance variation by a factor of 7 and improves their throughput by an average of 3.5 times when multiple applications share remote memory.Finally, we demonstrate that our insights can be generalized to accelerators and emerging AI workloads. We develop Concerto, a preemptive GPU runtime for large language model serving that harnesses idle GPU resources for offline inference tasks. By opportunistically batching offline inference tasks when online serving cannot fully saturate GPUs, Concerto significantly increases GPU utilization by an average of 2.35 times. By reactively preempting offline tasks upon online load bursts, Concerto reduces online serving latency by two orders of magnitude.Together, these systems form a new datacenter system stack that synergistically enhances performance, resource utilization, and cost efficiency, offering a transformative approach to modern datacenter management.
일반주제명  
Computer science
일반주제명  
Computer engineering
키워드  
Cloud computing
키워드  
Data centers
키워드  
Distributed parallel systems
키워드  
Machine learning
키워드  
Operating systems
키워드  
Scheduling resource management
기타저자  
University of California, Los Angeles Computer Science 0201
기본자료저록  
Dissertations Abstracts International. 86-03B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017164083
■00520250211152830
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798384082132
■035    ▼a(MiAaPQ)AAI31560455
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a004
■1001  ▼aQiao,  Yifan.
■24510▼aReinventing  Datacenter  System  Stacks  for  Resource  Harvesting
■260    ▼a[Sl]▼bUniversity  of  California,  Los  Angeles▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a233  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-03,  Section:  B.
■500    ▼aAdvisor:  Xu,  Harry  Guoqing;Kim,  Miryung.
■5021  ▼aThesis  (Ph.D.)--University  of  California,  Los  Angeles,  2024.
■520    ▼aThe  rise  of  cloud  computing  and  recent  AI  breakthroughs  have  radically  expanded  the  demand  for  datacenter  hardware  resources,  including  CPU,  memory,  and  accelerators  such  as  GPUs.  Despite  the  critical  need  to  improve  resource  utilization  and  reduce  operational  cost,  current  datacenter  system  stacks-comprising  OSes  and  runtime  systems-struggle  to  fully  utilize  hardware  resources  due  to  high  load  variability  and  stringent  performance  requirements  of  datacenter  workloads,  leading  to  substantial  waste  of  compute  and  memory  resources.This  dissertation  demonstrates  that  it  is  feasible  to  safely  and  efficiently  harvest  stranded  datacenter  resources,  even  when  they  are  intermittently  available  and  dispersed  across  servers.  Specifically,  we  identify  two  previously  overlooked  resource  harvesting  opportunities  in  today's  data  center  system  stacks.  First,  although  datacenter  applications  often  have  varying  and  potentially  large  resource  demands,  they  typically  include  elastic  components  that  can  be  safely  discarded  under  resource  pressure,  making  them  ideal  for  utilizing  idle  resources  with  temporal  availability.  Existing  operating  systems  and  runtime  systems,  though,  lack  proper  interfaces  for  applications  to  convey  such  semantics  and  take  advantage  of  idle  resources.  Second,  while  the  availability  of  resources  per  server  is  unpredictable,  combining  stranded  resources  across  servers  can  offer  better  overall  availability.  However,  this  opportunity  is  unavailable  to  many  datacenter  workloads  that  were  designed  for  running  on  a  single  machine.Driven  by  these  insights,  this  dissertation  rethinks  the  datacenter  system  stack  and  introduces  holistic  designs  for  OS  abstractions,  the  OS  kernel,  and  application  runtime  systems  for  resource  harvesting.  The  contributions  of  this  dissertation  are  fourfold.First,  we  investigate  how  to  harvest  resources,  especially  memory  which  is  inelastic  and  hard  to  re-assign  between  applications,  within  a  single  server.  We  introduce  Midas,  an  OS  memory  abstraction  that  allows  applications  to  use  idle  memory  for  storing  their  soft  state.  Midas  efficiently  manages  soft  memory  with  a  kernel-runtime  co-design,  achieving  near-optimal  performance  for  four  real-world  datacenter  applications  and  responding  to  extreme  memory  pressure  quickly  enough  to  avoid  running  out  of  memory.Second,  we  explore  how  to  harvest  resources  across  servers.  We  present  Hermit,  a  redesigned  OS  kernel  paging/swap  system  that  enables  applications  to  harvest  idle  memory  on  remote  servers  with  full  transparency  and  efficiency.  Hermit  allows  any  application  to  harness  remote  memory  without  changing  a  single  line  of  code,  making  it  practical  for  legacy  real-world  datacenter  applications.  It  also  achieves  three  orders  of  magnitude  lower  tail  latency  and  up  to  1.87  times  higher  throughput  for  latency-critical  and  batch-processing  applications,  respectively.Third,  built  atop  Hermit,  Canvas  is  a  resource  isolation  mechanism  for  the  kernel  swap  system  that  allows  multiple  applications  to  share  remote  memory  without  performance  interference.  By  segregating  resource  usages  and  access  patterns  of  co-running  applications,  Canvas  further  adaptively  optimizes  kernel  swap  for  each  application.  Our  evaluation  and  performance  study  with  a  wide  range  of  datacenter  applications  demonstrate  that  Canvas  reduces  performance  variation  by  a  factor  of  7  and  improves  their  throughput  by  an  average  of  3.5  times  when  multiple  applications  share  remote  memory.Finally,  we  demonstrate  that  our  insights  can  be  generalized  to  accelerators  and  emerging  AI  workloads.  We  develop  Concerto,  a  preemptive  GPU  runtime  for  large  language  model  serving  that  harnesses  idle  GPU  resources  for  offline  inference  tasks.  By  opportunistically  batching  offline  inference  tasks  when  online  serving  cannot  fully  saturate  GPUs,  Concerto  significantly  increases  GPU  utilization  by  an  average  of  2.35  times.  By  reactively  preempting  offline  tasks  upon  online  load  bursts,  Concerto  reduces  online  serving  latency  by  two  orders  of  magnitude.Together,  these  systems  form  a  new  datacenter  system  stack  that  synergistically  enhances  performance,  resource  utilization,  and  cost  efficiency,  offering  a  transformative  approach  to  modern  datacenter  management.
■590    ▼aSchool  code:  0031.
■650  4▼aComputer  science
■650  4▼aComputer  engineering
■653    ▼aCloud  computing
■653    ▼aData  centers
■653    ▼aDistributed  parallel  systems
■653    ▼aMachine  learning
■653    ▼aOperating  systems
■653    ▼aScheduling  resource  management
■690    ▼a0984
■690    ▼a0464
■690    ▼a0800
■71020▼aUniversity  of  California,  Los  Angeles▼bComputer  Science  0201.
■7730  ▼tDissertations  Abstracts  International▼g86-03B.
■790    ▼a0031
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17164083▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

미리보기

내보내기

chatGPT토론

Ai 추천 관련 도서


    신착도서 더보기
    최근 3년간 통계입니다.

    소장정보

    • 예약
    • 소재불명신고
    • 나의폴더
    • 우선정리요청
    • 비도서대출신청
    • 야간 도서대출신청
    소장자료
    등록번호 청구기호 소장처 대출가능여부 대출정보
    TF12000 전자도서 대출가능 마이폴더 부재도서신고 비도서대출신청 야간 도서대출신청

    * 대출중인 자료에 한하여 예약이 가능합니다. 예약을 원하시면 예약버튼을 클릭하십시오.

    해당 도서를 다른 이용자가 함께 대출한 도서

    관련 인기도서

    로그인 후 이용 가능합니다.