본문

서브메뉴

Resilient Cache Network Management: Algorithms, Analysis, Experiments
Resilient Cache Network Management: Algorithms, Analysis, Experiments
Resilient Cache Network Management: Algorithms, Analysis, Experiments

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20250211152110
ISBN  
9798384231783
DDC  
778.59
저자명  
Xie, Tian.
서명/저자  
Resilient Cache Network Management: Algorithms, Analysis, Experiments
발행사항  
[Sl] : The Pennsylvania State University, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
162 p
주기사항  
Source: Dissertations Abstracts International, Volume: 86-03, Section: B.
주기사항  
Advisor: He, Ting.
학위논문주기  
Thesis (Ph.D.)--The Pennsylvania State University, 2024.
초록/해제  
요약As we all know, the computer networks rely on cache to improve its performance. For example, the DNS caches resource records, the CDN caches content, and the SDN caches flow rules. However, on one hand, we know that caches are vulnerable to attacks such as cache overflow and cache pollution attack. The Existing defenses to protect the cache can be mainly divided into two parts, which are attack traffic detection, or using system-specific countermeasures. More importantly, all these mechanisms are external. Therefore, it remains to find out: how well can a cache to protect itself from attack. On the other hand, in-network caching and flexible routing are two of the most celebrated advantages of next generation network infrastructures. Yet few solutions are available for jointly optimizing caching and routing that provide performance guarantees for an arbitrary topology. In this thesis, we are going to explore different perspectives of cache network management, trying to make it resilient to mansion attacks, security threads, or naturally occurring faults.In our first piece of work, we zoom into a single cache node, improving its resilience by designing/picking the suitable replacement policy. We study this problem by characterizing the attack resilience of cache replacement policies under cache pollution attack, which is also a form of Denial of Service attack. To understand the cache performance under such attacks, we analyze a suite of representative replacement policies under the framework of TTL approximation in how well they preserve the hit ratios for legitimate users, while incorporating the delay for the cache to obtain a missing content. We further develop a scheme to adapt the cache replacement policy based on the perceived level of attack. Our analysis and validation on real traces show that although no single policy is resilient to all the attack strategies, suitably adapting the replacement policy can notably improve the attack resilience of the cache. Motivated by these results, we implement selected policies as well as policy adaptation in an open-source SDN switch to manage flow rule replacement, which is shown to notably improve its resilience to pollution attacks.As our second piece of work, we zoom out to the multi-hop cache network, for which the control plane is centralized. Suppose there is a centralized optimizer, as in a Software Defined Network (SDN) that oversees the storage of all the content, routing of all the requests(flows), our focus is how to jointly optimize caching and routing. We take a holistic approach towards this fundamental problem by analyzing its complexity in all the cases and developing polynomial-time algorithms with approximation guarantees in important special cases. We also reveal the fundamental challenge in achieving guaranteed approximation in the general case and propose an alternating optimization algorithm with good performance and fast convergence. Our algorithms have demonstrated superior performance in both routing cost and congestion compared to the state-of-the-art solutions in evaluations based on real topology and request traces.Shifting focus to cache vulnerabilities (in SDN), our third piece of work tackle the exposed nature of internal switch flow table sizes. Though previous work exposed edge switch manipulation, existing methods falter when navigating deeper within the network. Our work bridges this gap by unveiling the first solution to infer internal switch flow table sizes. Leveraging crafted probing sequences and robust statistical analysis, our Multi-Switch Size Inference (MSSI) algorithm estimates sizes along probing paths. This information feeds the Multi-Switch Size Matching (MSSM) algorithm, which pinpoints specific switches based on size intersections from multiple paths. Importantly, our solution relies solely on readily available Round Trip Time (RTT) measurements, making it broadly applicable without specialized network access. Rigorous theoretical analysis establishes accuracy bounds and correctness guarantees for MSSI and MSSM. Extensive Mininet experiments further validate our approach, showcasing its superior accuracy compared to state-of-the-art techniques.Our future research journey tackles diverse network challenges, from enhancing cache resilience and securing SDN to optimizing data transmission in dynamic environments. This future thrust focuses on the latter, specifically within the realm of overlay networks. These versatile networks play a critical role in modern data transmission, but their dynamic nature requires adaptable routing strategies for optimal performance. To address this dynamic challenge, we delve into the world of multi-arm bandits algorithms. These algorithms offer a promising framework for online shortest path routing in overlay networks. By leveraging the principles of exploration and exploitation, our proposed approach continuously evaluates and learns from real-time data, adjusting routing decisions based on the most up-to-date network conditions. This adaptive approach promises to improve efficiency and performance in overlay networks, ensuring the swift and seamless delivery of data even in the face of constant change.
일반주제명  
Video recordings
일반주제명  
Pollution
일반주제명  
Markov analysis
일반주제명  
Film studies
기타저자  
The Pennsylvania State University.
기본자료저록  
Dissertations Abstracts International. 86-03B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008250123s2024        us                              c    eng  d
■001000017162900
■00520250211152110
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798384231783
■035    ▼a(MiAaPQ)AAI31353813
■035    ▼a(MiAaPQ)PennState21247tbx5027
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a778.59
■1001  ▼aXie,  Tian.
■24510▼aResilient  Cache  Network  Management:  Algorithms,  Analysis,  Experiments
■260    ▼a[Sl]▼bThe  Pennsylvania  State  University▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a162  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  86-03,  Section:  B.
■500    ▼aAdvisor:  He,  Ting.
■5021  ▼aThesis  (Ph.D.)--The  Pennsylvania  State  University,  2024.
■520    ▼aAs  we  all  know,  the  computer  networks  rely  on  cache  to  improve  its  performance.  For  example,  the  DNS  caches  resource  records,  the  CDN  caches  content,  and  the  SDN  caches  flow  rules.  However,  on  one  hand,  we  know  that  caches  are  vulnerable  to  attacks  such  as  cache  overflow  and  cache  pollution  attack.  The  Existing  defenses  to  protect  the  cache  can  be  mainly  divided  into  two  parts,  which  are  attack  traffic  detection,  or  using  system-specific  countermeasures.  More  importantly,  all  these  mechanisms  are  external.  Therefore,  it  remains  to  find  out:  how  well  can  a  cache  to  protect  itself  from  attack.  On  the  other  hand,  in-network  caching  and  flexible  routing  are  two  of  the  most  celebrated  advantages  of  next  generation  network  infrastructures.  Yet  few  solutions  are  available  for  jointly  optimizing  caching  and  routing  that  provide  performance  guarantees  for  an  arbitrary  topology.  In  this  thesis,  we  are  going  to  explore  different  perspectives  of  cache  network  management,  trying  to  make  it  resilient  to  mansion  attacks,  security  threads,  or  naturally  occurring  faults.In  our  first  piece  of  work,  we  zoom  into  a  single  cache  node,  improving  its  resilience  by  designing/picking  the  suitable  replacement  policy.  We  study  this  problem  by  characterizing  the  attack  resilience  of  cache  replacement  policies  under  cache  pollution  attack,  which  is  also  a  form  of  Denial  of  Service  attack.  To  understand  the  cache  performance  under  such  attacks,  we  analyze  a  suite  of  representative  replacement  policies  under  the  framework  of  TTL  approximation  in  how  well  they  preserve  the  hit  ratios  for  legitimate  users,  while  incorporating  the  delay  for  the  cache  to  obtain  a  missing  content.  We  further  develop  a  scheme  to  adapt  the  cache  replacement  policy  based  on  the  perceived  level  of  attack.  Our  analysis  and  validation  on  real  traces  show  that  although  no  single  policy  is  resilient  to  all  the  attack  strategies,  suitably  adapting  the  replacement  policy  can  notably  improve  the  attack  resilience  of  the  cache.  Motivated  by  these  results,  we  implement  selected  policies  as  well  as  policy  adaptation  in  an  open-source  SDN  switch  to  manage  flow  rule  replacement,  which  is  shown  to  notably  improve  its  resilience  to  pollution  attacks.As  our  second  piece  of  work,  we  zoom  out  to  the  multi-hop  cache  network,  for  which  the  control  plane  is  centralized.  Suppose  there  is  a  centralized  optimizer,  as  in  a  Software  Defined  Network  (SDN)  that  oversees  the  storage  of  all  the  content,  routing  of  all  the  requests(flows),  our  focus  is  how  to  jointly  optimize  caching  and  routing.  We  take  a  holistic  approach  towards  this  fundamental  problem  by  analyzing  its  complexity  in  all  the  cases  and  developing  polynomial-time  algorithms  with  approximation  guarantees  in  important  special  cases.  We  also  reveal  the  fundamental  challenge  in  achieving  guaranteed  approximation  in  the  general  case  and  propose  an  alternating  optimization  algorithm  with  good  performance  and  fast  convergence.  Our  algorithms  have  demonstrated  superior  performance  in  both  routing  cost  and  congestion  compared  to  the  state-of-the-art  solutions  in  evaluations  based  on  real  topology  and  request  traces.Shifting  focus  to  cache  vulnerabilities  (in  SDN),  our  third  piece  of  work  tackle  the  exposed  nature  of  internal  switch  flow  table  sizes.  Though  previous  work  exposed  edge  switch  manipulation,  existing  methods  falter  when  navigating  deeper  within  the  network.  Our  work  bridges  this  gap  by  unveiling  the  first  solution  to  infer  internal  switch  flow  table  sizes.  Leveraging  crafted  probing  sequences  and  robust  statistical  analysis,  our  Multi-Switch  Size  Inference  (MSSI)  algorithm  estimates  sizes  along  probing  paths.  This  information  feeds  the  Multi-Switch  Size  Matching  (MSSM)  algorithm,  which  pinpoints  specific  switches  based  on  size  intersections  from  multiple  paths.  Importantly,  our  solution  relies  solely  on  readily  available  Round  Trip  Time  (RTT)  measurements,  making  it  broadly  applicable  without  specialized  network  access.  Rigorous  theoretical  analysis  establishes  accuracy  bounds  and  correctness  guarantees  for  MSSI  and  MSSM.  Extensive  Mininet  experiments  further  validate  our  approach,  showcasing  its  superior  accuracy  compared  to  state-of-the-art  techniques.Our  future  research  journey  tackles  diverse  network  challenges,  from  enhancing  cache  resilience  and  securing  SDN  to  optimizing  data  transmission  in  dynamic  environments.  This  future  thrust  focuses  on  the  latter,  specifically  within  the  realm  of  overlay  networks.  These  versatile  networks  play  a  critical  role  in  modern  data  transmission,  but  their  dynamic  nature  requires  adaptable  routing  strategies  for  optimal  performance.  To  address  this  dynamic  challenge,  we  delve  into  the  world  of  multi-arm  bandits  algorithms.  These  algorithms  offer  a  promising  framework  for  online  shortest  path  routing  in  overlay  networks.  By  leveraging  the  principles  of  exploration  and  exploitation,  our  proposed  approach  continuously  evaluates  and  learns  from  real-time  data,  adjusting  routing  decisions  based  on  the  most  up-to-date  network  conditions.  This  adaptive  approach  promises  to  improve  efficiency  and  performance  in  overlay  networks,  ensuring  the  swift  and  seamless  delivery  of  data  even  in  the  face  of  constant  change.
■590    ▼aSchool  code:  0176.
■650  4▼aVideo  recordings
■650  4▼aPollution
■650  4▼aMarkov  analysis
■650  4▼aFilm  studies
■690    ▼a0900
■690    ▼a0796
■71020▼aThe  Pennsylvania  State  University.
■7730  ▼tDissertations  Abstracts  International▼g86-03B.
■790    ▼a0176
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17162900▼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
    TF12840 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.