Codes can reduce queueing delay in data centers

Huang, Longbo; Pawar, Sameer; Zhang, Hao; Ramchandran, Kannan

doi:10.1109/isit.2012.6284026

Cited by 90 publications

(152 citation statements)

References 11 publications

(16 reference statements)

Supporting

Mentioning

147

Contrasting

Order By: Relevance

“…Related design can also be found in data access scheduling [14,16,19,20], access collision avoidance [17,18], and encoding/decoding time optimization [32,33] and there are also some work using the LT erasure codes to adjust the system to meet user requirements such as availability, integrity and confidentiality [6]. Restricting to the special case of a single file or homogeneous files, service delay bounds of erasure-coded storage have been recently studied in [40,45,49,50]. Queuing-theoretic analysis.…”

Section: B Related Workmentioning

confidence: 99%

“…For a single file or multiple but homogeneous files, under an assumption of exponential service time distribution, the authors in [5] proved an asymptotic result for symmetric large-scale systems which can be applied to provide a computable approximation for expected latency, however, under a assumption that chunk placement is fixed and so is coding policy for all requests, which is not the case in reality. Also, the authors in [40,45] proposed a block-onescheduling policy that only allows the request at the head of the buffer to move forward. An upper bound on the average latency of the storage system is provided through queuingtheoretic analysis for MDS codes with k = 2.…”

Section: B Related Workmentioning

confidence: 99%

“…[40,45,49,50], much less is known about the latency performance of multiple heterogeneous files that are coded with different parameters and share common storage servers. The main goal of this paper can be illustrated by an abstracted example shown in Fig.…”

Section: Introduction a Motivationmentioning

confidence: 99%

See 2 more Smart Citations

Joint latency and cost optimization for erasurecoded data center storage

Xiang

Lan

Aggarwal

et al. 2014

SIGMETRICS Perform. Eval. Rev.

View full text Add to dashboard Cite

Abstract-Modern distributed storage systems offer large capacity to satisfy the exponentially increasing need of storage space. They often use erasure codes to protect against disk and node failures to increase reliability, while trying to meet the latency requirements of the applications and clients. This paper provides an insightful upper bound on the average service delay of such erasure-coded storage with arbitrary service time distribution and consisting of multiple heterogeneous files. Not only does the result supersede known delay bounds that only work for a single file or homogeneous files, it also enables a novel problem of joint latency and storage cost minimization over three dimensions: selecting the erasure code, placement of encoded chunks, and optimizing scheduling policy. The problem is efficiently solved via the computation of a sequence of convex approximations with provable convergence. We further prototype our solution in an open-source, cloud storage deployment over three geographically distributed data centers. Experimental results validate our theoretical delay analysis and show significant latency reduction, providing valuable insights into the proposed latency-cost tradeoff in erasure-coded storage.

show abstract

Section: B Related Workmentioning

confidence: 99%

Section: B Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Joint latency and cost optimization for erasurecoded data center storage

Xiang

Lan

Aggarwal

et al. 2014

SIGMETRICS Perform. Eval. Rev.

View full text Add to dashboard Cite

show abstract

“…$15.00 DOI: 10.1145/nnnnnnn.nnnnnnn simultaneously download from di erent replicas, or each user can issue simultaneous requests to multiple replicas and wait for the rst download to nish [6,7,15]. Data access systems arising in erasure coded storage have received a lot of a ention for all-data (see for example [12,14,19,24] and references therein) and some for hot-data download [16,17], and usually involve single or multiple inter-dependent fork-join queueing systems.…”

Section: Introductionmentioning

confidence: 99%

Simplex Queues for Hot-Data Download

Aktaş

Najm

Soljanin

2017

Proceedings of the 2017 ACM SIGMETRICS / International Conference on Measurement and Modeling of Computer Systems

View full text Add to dashboard Cite

In cloud storage systems, hot data is usually replicated over multiple disks/servers in order to accommodate simultaneous access by multiple users as well as increase the fault tolerance of the system. Recent cloud storage research has proposed using availability codes, which is a special class of erasure codes, as a more storagee cient way to store hot data. ese codes enable data recovery from multiple, small disjoint groups of servers. e number of the recovery groups is referred to as the availability and the size of each group as the locality of the code. Up till now, we have very limited knowledge on how code locality and availability a ect data access time. Data download from these systems involves multiple fork-join queues operating in-parallel, making the analysis of access time a very challenging problem.In this paper, we present an analysis of average data access time in storage systems employing simplex codes, which are an important, in certain sense optimal, class of availability codes. We generalize the analysis for codes with locality 2 and any degree of availability. Speci cally, using a queueing theoretic approach, we derive bounds and approximations on the average response time for two di erent Poisson request arrival models. We also compare two scheduling strategies for reduced access time and load balancing.

show abstract

“…In [4], latency delay was reduced by choosing MDS codes of appropriate rates. Latency comparison between a simple replication scheme and MDS codes was pioneered by Huang et al [5] using queuing theory. It was shown that for k = 2, the average latency for serving a packet decreases significantly when a certain scheduling model is used.…”

Section: Introductionmentioning

confidence: 99%

Coding for Improved Throughput Performance in Network Switches

Cohen

Cassuto

2017

IEEE/ACM Trans. Networking

View full text Add to dashboard Cite

Network switches and routers need to serve packet writes and reads at rates that challenge the most advanced memory technologies. As a result, scaling the switching rates is commonly done by parallelizing the packet I/Os using multiple memory units. For improved read rates, packets can be coded upon write, thus giving more flexibility at read time to achieve higher utilization of the memory units. This paper presents a detailed study of coded network switches, and in particular how to design them to maximize the throughput advantages over standard uncoded switches. Toward that objective the paper contributes a variety of algorithmic and analytical tools to improve and evaluate the throughput performance. The most interesting finding of this study is that the placement of packets in the switch memory is the key to both high performance and algorithmic efficiency. One particular placement policy we call "design placement" is shown to enjoy the best combination of throughput performance and implementation feasibility.

show abstract

Codes can reduce queueing delay in data centers

Cited by 90 publications

References 11 publications

Joint latency and cost optimization for erasurecoded data center storage

Joint latency and cost optimization for erasurecoded data center storage

Simplex Queues for Hot-Data Download

Coding for Improved Throughput Performance in Network Switches

Contact Info

Product

Resources

About