I/O - Workload (Access Pattern)

Card Puncher Data Processing

About

Each server has unique workload characteristics. However there are some basic pattern that we can found in the storage demands.

Although storage system workloads differ quantitatively in terms of the:

  • transfer sizes,
  • locality,
  • intensity,
  • and distribution of reads vs. writes,

the most common applications can be reduced to six different measurement classifications.

By simulating these access patterns, system integrators and administrators can evaluate the performance capabilities for most server and workstation workloads prior to deployment.

I/O Access pattern I/O Characteristics Typical Applications
Streaming Reads 100% Reads; Large contiguous requests; 1-64 concurrent requests. May be threaded. Media Servers (Video on-demand, etc.). Virtual Tape Libraries (VTL), Application Servers
Streaming Writes 100% Writes; Large contiguous requests; 1-64 concurrent requests. May be threaded. Media Capture, VTL, Medical Imaging, Archiving, Backup, Video Surveillance, Reference Data
OLTP Typically 2KB – 16KB request sizes; Read modify, write, verify operations resulting in 2 reads for every write; Primarily random accesses. Large number of concurrent requests. When running SQL statements in parallel, Database will perform typically large random I/Os. Databases (SAP, Oracle, SQL), Online Transaction Servers
File Server Moderate distribution of request sizes from 4KB to 64KB, however 4KB and 64KB comprise 70% of requests; Primarily random; Generally 4 reads for every write operation. Large number of concurrent requests during peak operational periods. File and Printer Servers, e-mail (Exchange, Notes), Decision Support Systems
Web Server Wide distribution of request sizes from 512 bytes to 512KB; Primarily random accesses; Large number of concurrent requests during peak operational periods Web Services, Blogs, RSS Feeds, Shopping Carts, Search Engines, Storage Services
Workstations Primarily small to medium request sizes; 80% sequential and 20% random; Generally 4 reads for every write operation. 1-4 concurrent requests. Business Productivity, Scientific/Engineering Applications

The default Access Specification of IoMeter is a typical database workload: 2-Kilobyte random I/Os with a mix of 67% reads and 33% writes (2 reads for every write)

Type of physical disk Sustain throughput Read Access
HDDs 20-30 MB/s for large random reads
HDDs 100-110 MB/s for large sequential reads

1)





Discover More
Card Puncher Data Processing
Business - Online transaction processing (OLTP)

Online transaction processing (or OLTP) refers to a class of application that support a business transaction process. Databases (SAP, Oracle, SQL) supporting Online Transaction Processing The...
Card Puncher Data Processing
Data Storage - Data Striping (I/O parallelism)

In computer data storage, data striping is the technique of segmenting logically sequential data, such as a file, in a way that accesses of sequential segments are made to different physical storage devices....
Data System Architecture
Data Warehousing - The Workload is always mixed

Also known as: active data warehousing, operational data warehousing etc. All this indicate something similar, namely a diverse workload running on a data warehouse system concurrently: Whether...
Card Puncher Data Processing
I/O - (Input/Output|Read/Write) - Data Access

I/O devices can be interpreted as streams, as they produce or consume potentially unlimited data over time. IO = Input / Output = Writing and Reading data. It's an umbrella term that regroups IO transfer...
Card Puncher Data Processing
I/O - IOmeter (Workload Generator)

IOmeter is a disk-testing, IO workload generator tool which generate and measure storage performance. Iometer is pronounced “eye-OM-i-ter,” to rhyme with “thermometer.”...
Data Path From Disk To Cpu
IO - Data Path / Balanced System

The data paths is the path composed of all hardware components that are needed to get the data from: a storage device (generally disk drive or network drive) to the CPU It is important to understand...
Stream Vs Batch
Stream vs Batch

This article talks Stream Processing vs Batch Processing. The most important difference is that: in batch processing the size (cardinality) of the data to process is known whereas in a stream processing,...
Card Puncher Data Processing
What is Data Processing (Data Integration)?

Card puncher Data processing is a more general term for manipulating data whereas data integration is the integration...



Share this page:
Follow us:
Task Runner