TPC-C: The TPC-C benchmark is designed to measure a server’s ability to function as an on-line transaction processing (OLTP) database server. This benchmark simulates a complete computing environment where a population of users executes order-entry transactions against a database. These transactions include placing a new order, making or receiving a payment, checking the order status, monitoring delivery, and tracking stock levels. The TPC-C result is a measurement of the number of new order transactions generated per minute while the system is also executing the other four transaction types.
MMB2: The MAPI Messaging Benchmark (MMB2) was developed by Microsoft to measure a server’s ability to function as a Microsoft Exchange server platform. MMB2 measures the throughput of a “Medium User” executing common tasks such as sending, browsing, reading, and forwarding email messages, as well as scheduling tasks and using distribution lists over an eight-hour working day.
SPECweb®99: SPECweb99 is designed to measure a system's ability to act as a web server for static and dynamic web pages. In the benchmark set-up, a number of client systems create static and dynamic page requests against a server, simulating a real-world web server workload. The returned SPECweb99 benchmark score is the maximum number of conforming simultaneous connections a web server is able to support while still meeting specific throughput and error rate requirements. Conforming connections must be sustained at a specified maximum bit rate with a maximum segment size. This is intended to realistically model conditions that will be seen on the Internet during the lifetime of this benchmark.
SPECweb99_SSL: Like SPECweb99, SPECweb99_SSL is designed to measure a system’s ability to act as a web server for static and dynamic pages. SPECweb99_SSL differs, however, in that pages are requested using HTTP over the Secure Sockets Layer Protocol (HTTPS) and the returned benchmark result is intended to measure the system’s ability to act as a secure web server.
SPECjbb™2000: SPECjbb2000 is designed to measure a system’s ability to act as a middle-tier java application server in a 3-tier system. A 3-tier system consists of a client (tier 1), a Java application server (tier 2), and a database (tier 3). The Java application server contains the business logic and object caches that manage incoming requests from the client, retrieve the appropriate information from the database, and return that information to the client. Java application servers are commonly used in ERP, CRM, e-Business, and other tiered applications in which a web browser is used to access information contained in a database. The SPECjbb2000 benchmark is modeled around a wholesale company with warehouses that serve a number of districts. The SPECjbb2000 result measures the throughput of the underlying Java platform, which is the rate at which certain business operations are performed per second.
SPECint®2000: SPECint2000 is designed to measure and compare compute-intensive integer performance between systems. This benchmark emphasizes the performance of:
The processor (CPU)
The memory architecture
The compilers
SPECint2000 consists of twelve integer benchmarks developed from actual end-user applications. Integer-heavy applications are most common within IT departments and enterprise server deployments. Applications such as databases, e-mail servers, Java application servers, and web servers all typically perform better when running on a processor with excellent integer performance.
SPECint_rate™2000: This benchmark runs the same algorithms used in SPECint2000, but it executes multiple instances of the benchmark at one time (typically one instance per processor in the system). SPECint_rate2000 measures a system’s ability to carry out multiple compute-intensive integer operations at one time. This benchmark is designed to measure a multi-processor system’s ability to scale well while running integer-based applications such as database servers, email servers, or web servers.
SPECfp®2000: SPECfp2000 is designed to measure and compare compute-intensive floating-point performance between systems. This benchmark emphasizes the performance of:
The processor (CPU)
The memory architecture
The compilers
SPECfp2000 consists of fourteen floating-point benchmarks developed from actual end-user applications. Intense floating-point applications are most common within engineering and research environments. Applications such as computational fluid dynamics, CAD/CAM, digital content creation (DCC), rendering, and financial modeling tools typically perform better when running on a processor with excellent floating-point performance.
SPECfp_rate2000: This benchmark runs the exact same algorithms used in SPECfp2000, but it executes multiple instances of the benchmark at one time (typically one instance per processor in the system). SPECfp_rate2000 measures a system’s ability to carry out multiple compute-intensive floating-point operations at one time. This benchmark is designed to measure a multi-processor system’s ability to scale well while running floating-point-based applications such as CAD/CAM, DCC, and other scientific applications.
