KV-Cache: A Scalable High-Performance Web-Object Cache for Manycore

Waddington, Daniel; Colmenares, Juan A.; Kuang, Jilong; Song, Fengguang

doi:10.1109/ucc.2013.34

Cited by 8 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, many websites have highly-personalized content, thus rendering wholepage web caches is mostly useless; application-level caches can be used to separate shared content from customized content, and then the shared content can be cached and reused among users [4]. Memcached [26] and Redis [27] are popular solutions and are a critical web infrastructure component for some big players such as Amazon, Facebook, LinkedIn, Twitter, Wikipedia, and YouTube [28].…”

Section: Background and Related Workmentioning

confidence: 99%

A Qualitative Study of Application-Level Caching

Mertz

Nunes

2017

IIEEE Trans. Software Eng.

View full text Add to dashboard Cite

Latency and cost of Internet-based services are encouraging the use of application-level caching to continue satisfying users' demands, and improve the scalability and availability of origin servers. Despite its popularity, this level of caching involves the manual implementation by developers and is typically addressed in an ad-hoc way, given that it depends on specific details of the application. As a result, application-level caching is a time-consuming and error-prone task, becoming a common source of bugs. Furthermore, it forces application developers to reason about a crosscutting concern, which is unrelated to the application business logic. In this paper, we present the results of a qualitative study of how developers handle caching logic in their web applications, which involved the investigation of ten software projects with different characteristics. The study we designed is based on comparative and interactive principles of grounded theory, and the analysis of our data allowed us to extract and understand how developers address cache-related concerns to improve performance and scalability of their web applications. Based on our analysis, we derived guidelines and patterns, which guide developers while designing, implementing and maintaining application-level caching, thus supporting developers in this challenging task that is crucial for enterprise web applications.

show abstract

Section: Background and Related Workmentioning

confidence: 99%

A Qualitative Study of Application-Level Caching

Mertz

Nunes

2017

IIEEE Trans. Software Eng.

View full text Add to dashboard Cite

show abstract

“…Indeed, this is true for I/O intensive applications due to limitations in I/O devices and critical sections in device drivers [6,10,21], but this argument is rarely true for computation-intensive applications. Computation-intensive applications typically perform in-memory computations and can minimize the overhead of system calls.…”

Section: Introductionmentioning

confidence: 99%

Scaling up matrix computations on shared-memory manycore systems with 1000 CPU cores

Song

Dongarra

2014

Proceedings of the 28th ACM International Conference on Supercomputing

Self Cite

View full text Add to dashboard Cite

While the growing number of cores per chip allows researchers to solve larger scientific and engineering problems, the parallel efficiency of the deployed parallel software starts to decrease. This unscalability problem happens to both vendorprovided and open-source software and wastes CPU cycles and energy. By expecting CPUs with hundreds of cores to be imminent, we have designed a new framework to perform matrix computations for massively many cores. Our performance analysis on manycore systems shows that the unscalability bottleneck is related to Non-Uniform Memory Access (NUMA): memory bus contention and remote memory access latency. To overcome the bottleneck, we have designed NUMA-aware tile algorithms with the help of a dynamic scheduling runtime system to minimize NUMA memory accesses. The main idea is to identify the data that is, either read a number of times or written once by a thread resident on a remote NUMA node, then utilize the runtime system to conduct data caching and movement between different NUMA nodes. Based on the experiments with QR factorizations, we demonstrate that our framework is able to achieve great scalability on a 48-core AMD Opteron system (e.g., parallel efficiency drops only 3% from one core to 48 cores). We also deploy our framework to an extreme-scale shared-memory SGI machine which has 1024 CPU cores and runs a single Linux operating system image. Our framework continues to scale well, and can outperform the vendoroptimized Intel MKL library by up to 750%.

show abstract