Workload characterization of interactive cloud services on big and small server platforms

Chen, Shuang; Gal-On, Shay; Delimitrou, Christina; Manne, Srilatha; Martínez, José F.

doi:10.1109/iiswc.2017.8167770

Cited by 29 publications

(13 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, we quantify how effective current datacenter architectures are at running microservices, as well as how datacenter hardware needs to change to better accommodate their performance and resource requirements (Section 4). This includes analyzing the cycle breakdown in modern servers, examining whether big or small cores are preferable [25,35,41,42,[46][47][48], determining the pressure microservices put on instruction caches [37,52], and exploring the potential they have for hardware acceleration [24,27,38,49,71]. We show that despite the small amount of computation per microservice, the latency requirements of each individual tier are much stricter than for typical applications, putting more pressure on predictably high single-thread performance.…”

Section: Cluster Management Implicationsmentioning

confidence: 99%

“…Brawny vs. wimpy cores: There has been a lot of work on whether small servers can replace high-end platforms in the cloud [25,[46][47][48]. Despite the power benefits of simple cores, interactive services still achieve better latency in servers that optimize for single-thread performance.…”

Section: Architectural Implicationsmentioning

confidence: 99%

“…Apart from frequency scaling, there are platforms designed with low-power cores to begin with. We also evaluate the end-to-end services on two Cavium Thun-derX boards (2 sockets, 48 in-order cores per socket, 1.8GHz each, and a 16-way shared 16MB LLC) [25]. The boards are connected on the same ToR switch as the rest of our cluster, and their memory and network subsystems are the same as the other servers.…”

Section: Architectural Implicationsmentioning

confidence: 99%

See 2 more Smart Citations

An Open-Source Benchmark Suite for Microservices and Their Hardware-Software Implications for Cloud & Edge Systems

Gan

Zhang

Cheng

et al. 2019

Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Syst

Self Cite

376

245

View full text Add to dashboard Cite

Cloud services have recently started undergoing a major shift from monolithic applications, to graphs of hundreds of loosely-coupled microservices. Microservices fundamentally change a lot of assumptions current cloud systems are designed with, and present both opportunities and challenges when optimizing for quality of service (QoS) and utilization. In this paper we explore the implications microservices have across the cloud system stack. We first present Death-StarBench, a novel, open-source benchmark suite built with microservices that is representative of large end-to-end services, modular and extensible. DeathStarBench includes a

show abstract

Section: Cluster Management Implicationsmentioning

confidence: 99%

Section: Architectural Implicationsmentioning

confidence: 99%

Section: Architectural Implicationsmentioning

confidence: 99%

See 1 more Smart Citation

An Open-Source Benchmark Suite for Microservices and Their Hardware-Software Implications for Cloud & Edge Systems

Gan

Zhang

Cheng

et al. 2019

Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Syst

Self Cite

376

245

View full text Add to dashboard Cite

show abstract

“…to handle network interrupts across the server socket when operating at max load. Allowing LC threads to share cores with IRQ cores leads to both lower throughput and higher latency [16]. 8GB of memory is exclusively allocated to the OS.…”

Section: Characterizationmentioning

confidence: 99%

Parties

Chen

Delimitrou

Martínez

2019

Proceedings of the Twenty-Fourth International Conference on Architectural Support for Programming Languages and Operating Syst

Self Cite

161

View full text Add to dashboard Cite

Multi-tenancy in modern datacenters is currently limited to a single latency-critical, interactive service, running alongside one or more low-priority, best-efort jobs. This limits the eiciency gains from multi-tenancy, especially as an increasing number of cloud applications are shifting from batch jobs to services with strict latency requirements. We present PARTIES, a QoS-aware resource manager that enables an arbitrary number of interactive, latency-critical services to share a physical node without QoS violations. PARTIES leverages a set of hardware and software resource partitioning mechanisms to adjust allocations dynamically at runtime, in a way that meets the QoS requirements of each co-scheduled workload, and maximizes throughput for the machine. We evaluate PARTIES on state-of-the-art server platforms across a set of diverse interactive services. Our results show that PARTIES improves throughput under QoS by 61% on average, compared to existing resource managers, and that the rate of improvement increases with the number of co-scheduled applications per physical host. CCS Concepts • Computer systems organization → Cloud computing; Real-time system architecture.

show abstract

“…We evaluate the end-to-end service on two Cavium ThunderX boards (2 sockets, 48 1.8GHz inorder cores per socket, and a 16-way shared 16M LLC). The boards are connected on the same ToR switch as the rest of our cluster, and their memory, network, and OS configurations are the same as the other servers [5]. Fig.…”

Section: Movie Streamingmentioning

confidence: 99%

The Architectural Implications of Cloud Microservices

Gan

Delimitrou

2018

IEEE Comput. Arch. Lett.

Self Cite

View full text Add to dashboard Cite

Cloud services have recently undergone a shift from monolithic applications to microservices, with hundreds or thousands of loosely-coupled microservices comprising the end-to-end application. Microservices present both opportunities and challenges when optimizing for quality of service (QoS) and cloud utilization. In this paper we explore the implications cloud microservices have on system bottlenecks, and datacenter server design. We first present and characterize an end-to-end application built using tens of popular open-source microservices that implements a movie renting and streaming service, and is modular and extensible. We then use the end-to-end service to study the scalability and performance bottlenecks of microservices, and highlight implications they have on the design of datacenter hardware. Specifically, we revisit the long-standing debate of brawny versus wimpy cores in the context of microservices, we quantify the I-cache pressure they introduce, and measure the time spent in computation versus communication between microservices over RPCs. As more cloud applications switch to this new programming model, it is increasingly important to revisit the assumptions we have previously used to build and manage cloud systems.

show abstract

Workload characterization of interactive cloud services on big and small server platforms

Cited by 29 publications

References 31 publications

An Open-Source Benchmark Suite for Microservices and Their Hardware-Software Implications for Cloud & Edge Systems

An Open-Source Benchmark Suite for Microservices and Their Hardware-Software Implications for Cloud & Edge Systems

Parties

The Architectural Implications of Cloud Microservices

Contact Info

Product

Resources

About