Benchmarking cloud serving systems with YCSB

By Brian F. Cooper, Adam Silberstein, Erwin Tam, Raghu Ramakrishnan, Russell Sears. ACM Symposium on Cloud Computing, 2010

4,215 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Information Systems

View Paper: https://www.semanticscholar.org/paper/c69859fd9ee74e5287c5dbf22ee7e82663fe8bdd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Information Systems

Citations in the corpus

Next generation cloud computing: New trends and research directions (2017, methodology): “Performance is measured on the cloud using a variety of techniques, such as benchmarking to facilitate the selection of resources that maximise performance of an application and periodic monitoring of the resources to ensure whether user-defined service level objectives are achieved [102, 103, 104, ”
Cache craftiness for fast multicore key-value storage (2012, background): “Fourth, a complete design that addresses all bottlenecks in the way of million-query-per-second performance.”
MemC3: Compact and Concurrent MemCache with Dumber Caching and Smarter Hashing (2013, neutral): “Workload This experiment uses the same workload as in Section 5.3, with 95% GET s and 5% SET s generated by YCSB with zipf distribution.”
Consistent and Durable Data Structures for Non-Volatile Byte-Addressable Memory (2011, methodology): “Using the Yahoo Cloud Serving Benchmark [15], we observed that Tembo increases throughput by up to 250%–286% when compared to memory-backed Cassandra, a two-level data store.”
CockroachDB: The Resilient Geo-Distributed SQL Database (2020, neutral): “CRDB’s support for hybrid deployments 2We use the official YCSB generator[76] with Spanner and JDBC clients.”
Scalable SQL and NoSQL data stores (2011, background): “In fact, BigTable, memcached, and Amazon's Dynamo [2] provided a " proof of concept " that inspired many of the data stores we describe here: • Memcached demonstrated that in-memory indexes can be highly scalable, distributing and replicating objects over multiple nodes.”
NV-Tree: Reducing Consistency Cost for NVM-based Single Level Systems (2015, methodology): “Finally, we implement a database interface layer to extend YCSB to support NV-Store to facilitate our performance evaluation.”
Performance Isolation and Fairness for Multi-Tenant Cloud Storage (2012, positive): “On the clients, we use the Yahoo Cloud Storage Benchmark (YCSB) [7] to generate a Zipf distributed key-value request workload (α = 0.99).”
Using One-Sided RDMA Reads to Build a Fast, CPU-Efficient Key-Value Store (2013, methodology): “We use the YCSB [3] benchmark to generate our workloads.”
The Case for Evaluating MapReduce Performance Using Workload Suites (2011, methodology): “Our approach differs from that taken by benchmarks that focus on IO byte stream properties [27].”
This paper is included in the Proceedings of the 12th USENIX Symposium on Operating Systems Design and Implementation (OSDI ’16). (positive): “We use workloads A to D from the YCSB benchmark suite [14].”
PARTIES: QoS-Aware Resource Partitioning for Multiple Interactive Services (2019, methodology): “For NGINX and MongoDB, we modiied popular open-source generators, wrk2 [5] and YCSB [17], from closed- to open-loop.”
SILT: a memory-efficient, high-performance key-value store (2011, methodology): “Workload Generation We use YCSB [17] to generate a key-value workload.”
Megastore: Providing Scalable, Highly Available Storage for Interactive Services (2011, background): “Open source database systems such as MySQL [10] do not scale up to the levels we require [17], while expensive commercial database systems like Oracle [4] significantly increase the total cost of ownership in large deployments in the cloud.”