Overview of Server Benchmark Standards: TPC and SPEC
The article explains the origins, metrics, and test suites of TPC and SPEC benchmarks, describes their various models for CPU, web, HPC and storage performance, shows how to query official results, and notes a promotional bundle of technical e‑books.
In the 1990s the Transaction Processing Performance Council (TPC) introduced the tpmC metric to evaluate online transaction processing capacity, simulating MIS and ERP workloads.
The Standard Performance Evaluation Corporation (SPEC) provides a range of benchmark suites such as SPECcpu2000/2006, SPECweb2005, SPEC HPC2002, SPEC MPI2006, SPECjAppServer2004, and SPECjbb2005, covering CPU, web servers, high‑performance computing, and Java applications.
Common benchmark families include high‑performance computing (Linpack), OLTP (TPC‑C), web services (SPECweb2005, TPC‑W), and Java application servers (SPECjbb2005), as well as vendor‑specific tests for Oracle and SAP.
SPEC’s CPU benchmarks are divided into base (e.g., SPECint_base2006) and peak (e.g., SPECint_2006) tests, with speed tests measuring single‑thread latency and rate tests measuring throughput under concurrent load.
The SPECsfs2008 file‑system benchmark assesses NAS throughput and latency, and high‑OPS results (e.g., 5 million OPS) enable massive workloads such as browsing millions of images per second.
To retrieve official SPEC results, users log into the SPEC/OSG Result Search Engine, select the desired configuration, filter by hardware or release date, and download the report.
TPC’s benchmark suite (TPC‑C, TPC‑H, TPC‑DS) defines the tpmC metric, relates it to SPECint_rate_base, and provides price‑performance ratios ($/tpmC) for server procurement in industries such as banking, finance, insurance, telecom, and government.
The article concludes with a promotional offer for a bundled collection of technical e‑books covering topics from RDMA and storage to cloud native and Kubernetes.
Architects' Tech Alliance
Sharing project experiences, insights into cutting-edge architectures, focusing on cloud computing, microservices, big data, hyper-convergence, storage, data protection, artificial intelligence, industry practices and solutions.
How this landed with the community
Was this worth your time?
0 Comments
Thoughtful readers leave field notes, pushback, and hard-won operational detail here.