EMC ECS Node Disk Performance Test

Nodes are the primary building block of the EMC ECS storage system. Each node has an array of disks attached which are managed by that node. The number of disks attached to a node depends on the performance requirement of the system.

Disk performance is one of the most important considerations for any storage system. The administrators have to be sure that all disks attached to each node are working to their optimal performance, any issue with any disk which may hamper read, write, or replication of data, may cause delays in response times or even loss of data. This is the reason each disk has to be monitored for performance and any indication of the potential problem is handled before it can cause any issue system performance.

Processing bottlenecks in disks can significantly delay reading from and/or writing to the disks, affecting user experience with the storage system as a whole. To prevent this, administrators need to keep a close watch on the I/O activity on the disks of the storage system, detect a slowdown well before the users notice the slowdown, and initiate necessary actions at the earliest. The EMC ECS Node Disk Performance test can help administrators to achieve this. This test auto-discovers disks on the nodes, and for each disk, reports the rate at which read and write operations were performed, using which any abnormalities can be detected before users start complaining of slowdowns.

This test focuses on the performance of each disk attached to a particular node and collects performance-related metrics. The most important measure it collects is the read-write latency.

Target of the test : A Dell EMC Elastic Cloud Storage System

Agent deploying the test : A remote agent

Outputs of the test : One set of results for each disk attached to the node.

Configurable parameters for the test
Parameter Description

Test period

How often should the test be executed .

Host

The host for which the test is to be configured. Since the storage device is managed using the IP address of its storage controller, the same will be displayed as host.

Port

The port number at which the specified host listens. By default, this is NULL.

ECS REST API Port

This is the port at which REST API connectivity is provided. By default, port 4443 is used.

Username and Password

To collect performance metrics from the target storage device, the eG agent should be configured with the credentials of a user who is vested with "read-only" privileges to access REST API of the target storage device. Specify the credentials of such a user in the Username and Password text boxes.

Confirm Password

Confirm the password by retyping it here.

Timeout Seconds

Specify the time duration for which this test should wait for a response from the storage system in the Timeout text box. By default, this is 60 seconds.

Measurements made by the test
Measurement Description Measurement Unit Interpretation

Erasure encoding reads

Indicates the rate at which disk bandwidth is used in system erasure coding read operations.

MB/Sec

Compare the rate with previous measurements or an optimal rate recorded. If the rate is below threshold or decreasing steadily over time, it may be an indication of a problem.

Consistency checker reads

Indicates the rate at which disk bandwidth is used to check for inconsistencies between protected data and its replicas.

MB/Sec

Compare the rate with previous measurements or an optimal rate recorded. If the rate is below threshold or decreasing steadily over time, it may be an indication of a problem.

XOR reads

Indicates the rate at which disk bandwidth is used in the system's XOR data protection operations. Note that XOR operations occur for systems with three or more sites (VDCs).

MB/Sec

Compare the rate with previous measurements or an optimal rate recorded.If the rate is below threshold or decreasing steadily over time, it may be an indication of a problem.

User reads

Total number of User using reads bandwidth

MB/Sec

Compare the rate with previous measurements or an optimal rate recorded. If the rate is below threshold or decreasing steadily over time, it may be an indication of a problem.

Geo reads

Rate at which disk bandwidth is used to support geo-replication operations.

MB/Sec

Compare the rate with previous measurements or an optimal rate recorded.If the rate is below threshold or decreasing steadily over time, it may be an indication of a problem.

Hardware recovery reads

Indicates the rate at which disk bandwidth is used to recover data after a hardware failure.

MB/Sec

Compare the rate with previous measurements or an optimal rate recorded.If the rate is below threshold or decreasing steadily over time, it may be an indication of a problem.

Total reads

Indicates the rate of total reads.

MB/Sec

Compare the rate with previous measurements or an optimal rate recorded. If the rate is below threshold or decreasing steadily over time, it may be an indication of a problem.

Erasure encoding writes

Indicates the rate at which disk bandwidth is used in system erasure coding operations.

MB/Sec

Compare the rate with previous measurements or an optimal rate recorded. If the rate is below threshold or decreasing steadily over time, it may be an indication of a problem.