SPC-1 SPC-1E V1.14 Official Specifications

目录

SPC BENCHMARK 1TM (SPC-1TM)

SPC BENCHMARK 1/ENERGYTM EXTENSION (SPC-1/ETM)

OFFICIAL SPECIFICATION

Revision 1.14 - Effective May 12, 2013

"The First Industry Standard Storage Benchmark"

1 CLAUSE0: Introduction

1.1 Preamble/前言

The SPC Benchmark-1TM (SPC-1) is a sophisticated performance measurement workload for storage subsystems. The benchmark simulates the demands placed upon on-line, non-volatile storage in a typical server class computer system. SPC-1 provides measurements in support of real world environments characterized by:

  • Demanding total I/O throughput requirements.
  • Sensitive I/O response time constraints.
  • Dynamic workload behaviors.
  • Substantial storage capacity requirements.
  • Diverse user populations and expectations.
  • Data persistence requirements to ensure preservation of data without

corruption or loss.

SPC基准1TM (SPC-1)是一个存储子系统复杂的性能测量工作负载。该基准模拟在线市场 需求

2 CLAUSE1: Workload Environments

2.1 Business and Application Environment

SPC-1 is comprised of a set of I/O operations designed to demonstrate the performance of a storage subsystem while performing the typical functions of a business critical application. SPC-1 represents a segment of applications characterized by predominately random I/O operations and requiring both queries as well as update operations (for example: OLTP systems, database systems, or mail server applications).

2.2 High-Level Workload Model/

The segment of applications represented by SPC-1 covers a broad range of user profiles, business functions and system configurations. Since the focus of SPC-1 is on the commonalties of those applications (e.g., high reliance on stored data, multi-user access, etc.), it was necessary to develop a model that would simplify the workload to the point that highlighted the similarities of its business segment while removing any conflicts and details that weren’t central to performance evaluation. The model used in SPC-1 has two central scaling components:

  • Business Scaling Units (BSUs)
  • Application Storage Units (ASUs)

2.3 Business Scaling Units (BSUs)

Business Scaling Units (BSUs) are the benchmark’s representation of an application’s user population. Each BSU represents the aggregate IO load created by a specified number of users. By focusing the benchmark on this aggregated IO load, SPC-1 is able to provide a scalable stimulus for the tested system that will provide a broad test of the storage configuration without getting lost in the detail that would be necessary for the accurate modeling of any one application. The result will be a workload that will retain its relevance across many generations of a particular application and through a broad spectrum of possible applications.

SPC-1 will be scaled by increasing or decreasing the number of BSUs. A more detailed technical description of BSUs may be found in Clause 4.

2.4 Application Storage Units (ASUs)

In the same way that the BSU generalizes the IO load presented to a tested system by an application, Application Storage Units (ASUs) are used to abstract the storage configuration that must respond to that IO load. An ASU is the logical entity identified by the application as the destination or source of data that requires persistence beyond the execution of the application. If the BSU can be thought of as the source of the benchmark workload, then the ASU becomes the sink, providing the logical environment in which the abstracted workload is run.

An ASU represents an abstraction of storage media and does not require a particular physical implementation. The physical implementation is determined by the Test Sponsor and must meet the storage configuration requirements stated in Clause 3.1.1. See Clause 4.3 for examples of supported configurations.

3 CLAUSE2: Data Repository

3.1 SPC-1 Storage Hierarchy

The SPC-1 data repository segments storage components into five distinct roles:

  • Physical Storage Capacity (PSC) defined in Clause 2.2.
  • Configured Storage Capacity (CSC) defined in Clause 2.3.
  • Addressable Storage Capacity (ASC) defined in Clause 2.4.
  • Logical Volumes (LV) defined in Clause 2.5.
  • Application Storage Unit (ASU) Capacity defined in Clause 2.6.

The relationship between the different storage capacities is illustrated in Figure 2-1. Included in the above storage capacities are:

  • Storage required for data protection.
  • Required Storage/Spares defined in Clause 2.3.2.
  • Global Storage Overhead defined in Clause 2.2.3.
  • Unused Storage defined in Clauses 2.2.4, 2.3.2, and 2.4.3.

3.2 Physical Storage Capacity (PSC)

3.2.1 Physical Storage Capacity is typically the formatted capacity of all Storage Devices that are physically present in the Tested Storage Configuration.

  1. In cases where the formatted capacity of a configured Storage Device is not

publicly available information, the reported value will be the capacity reported as available for application use.

  1. In cases where both the formatted capacity and the capacity available for

application use are publicly available information, the Test Sponsor will report either value. In such cases, the choice made by the Test Sponsor must be applied to all configured Storage Devices of the same model.

  1. If the Test Sponsor reconfigures the capacity of a Storage Device as

documented in Clauses 2.2.1.1 or 2.2.1.2 so that the resultant capacity is less than the original value, the difference is reported as Global Storage Overhead, as described in 2.2.3, and included in the reported Physical Storage Capacity.

3.2.2 All physical storage present in the TSC must be included in Physical Storage Capacity, whether or not it is cabled in or powered up.

3.2.3 Global Storage Overhead consists of the Physical Storage Capacity that is required for system use, such as metadata, and unavailable for use by application programs such as the SPC-1 Workload Generator.

4 CLAUSE3: Workload and I/O Operation Profile

5 CLAUSE4: Benchmark Configuration (BC), Tested Storage Configuration (TSC), and Workload Generator

6 CLAUSE5: Test Measurement Requirements (Execution Rules)

7 CLAUSE6: Data Persistence Requirements and Test

8 CLAUSE7: Reported Data

9 CLAUSE8: Pricing

10 CLAUSE9: Full Disclosure(信息披露)

11 CLAUSE10: Measurement, Audit, and Results Submission

12 CLAUSE11: Energy Extension

13 Appendix A: Letter of Good Faith Template

14 Appendix B: The Hierarchical Reuse Random Walk