Author: Bruce Mackenzie-Low

Xperf Rocks Part 2: Analyzing Storage Performance Data & Generating Graphs

by Bruce Mackenzie-Low

Bruce Mackenzie-Low covers Xperf, one of Microsoft’s hottest diagnostic tools. This article continues where "Xperf Rocks Part 1: Troubleshooting Storage Performance Problems" left off. Part 2 focuses on how to use Xperf to analyze event data and generate graphs & tables. Powerful graph options are explained, which allow you to quickly pinpoint storage bottlenecks, along with the detailed data in a tabular format.

Windows GPT Disks – Is Bigger Really Better?

with 14 Comments by Bruce Mackenzie-Low

Bruce Mackenzie-Low examines the benefits and disadvantages of MBR and GPT-based Windows disks. The 2 approaches differ in how they track the mapping of physical disk sectors to logical block numbers. MBR (Master Boot Record) is widely accepted, but partition size is limited to 2 terabytes. GPT (GUID Partition Table) was created to accommodate the larger partition sizes and offers greater resilience to corruption.

Leveraging the Cluster Validation Wizard for Troubleshooting Storage Problems

with 2 Comments by Bruce Mackenzie-Low

How to use the Cluster Validation Wizard, also known as Validate, to perform a variety of tests to ensure that cluster components are accurately configured and supported in a clustered environment. Bruce Mackenzie-Low explains using the Cluster Validation Wizard to systematically test the storage subsystem in order to isolate any failing components, including generation of a Validation Report which documents the tests and their results, along with hyperlinks to detailed troubleshooting information such as failing disks and server names.

Understanding PAL Reports to Identify Windows Storage Bottlenecks

with 1 Comment by Bruce Mackenzie-Low

How-to by Bruce Mackenzie-Low on using PAL (Performance Analysis of Logs) reports for highlighting which counters have exceeded predetermined thresholds. The lastest versions of PAL 2.0, now using PowerShell instead of VBScript, can simplify the troubleshooting of performance issues by identifying when counters have exceeded various thresholds, graphing their trends and identifying bottlenecks.