Deduplication on an XtremIO array is a process where redundant data blocks are identified and only

unique data is stored. This results in a reduction of the physical storage required, which is referred to

as the deduplication ratio. As the deduplication ratio increases, meaning more redundant data is

eliminated, the bandwidth effectively increases because less data needs to be transferred over the

network. Additionally, latency decreases because the system has to process fewer data blocks,

leading to faster response times1.

Reference:

The Dell EMC XtremIO X2: Next-Generation All-Flash Array document explains that XtremIO’s in-

memory deduplication is global across the entire cluster, which means that only unique data is

written to the SSDs. This inline deduplication not only saves significant capacity but can also improve

performance1.

The Introduction to XtremIO X2 Storage Array document details how the XtremIO X2 Storage Array

automatically reduces (deduplicates and compresses) data on the fly, as it enters the system, which

reduces the amount of data written to flash, improving the longevity of the media and driving down

cost2.

The Dell XtremIO X2 Storage Array uses an inline data deduplication process which involves

fingerprinting data blocks. The fingerprinting process is a part of the data reduction technique that

helps in identifying duplicate data blocks. When data enters the system, it is divided into small

chunks, and each chunk is fingerprinted using a hashing algorithm. The size of the fingerprint is

crucial as it determines the efficiency and accuracy of the deduplication process.

The specific bit size of the fingerprint created by the algorithm used by XtremIO is 256 bits. This

information is derived from the detailed descriptions of the system’s architecture and operation as

provided in the Dell EMC XtremIO X2 Storage Array documentation1. The document outlines the

system features, including inline data reduction techniques like deduplication and compression,

which are essential components of XtremIO’s data management capabilities.

The 256-bit fingerprint size ensures a balance between deduplication efficiency and the probability

of hash collisions (where different data blocks could result in the same fingerprint). A larger

fingerprint size would reduce the chance of collisions but would require more storage space for

metadata, while a smaller size would save metadata space but increase the risk of collisions.

Therefore, the 256-bit size is a strategic choice for the XtremIO system’s deduplication process.

In summary, the fingerprint bit size for XtremIO’s deduplication algorithm is 256 bits, which is

designed to optimize the system’s performance and data reduction capabilities while maintaining

data integrity.

When entering a generic workload into the XtremIO Sizing Tool, the required values are IOPs, I/O

size, and bandwidth. These parameters are essential for accurately sizing the storage system to

ensure it can handle the expected workload. IOPs (Input/Output Operations Per Second) measure

the number of read/write operations the storage system can perform in a second. I/O size refers to

the size of each read/write operation, and bandwidth is the total amount of data that can be

transferred in a given time frame.

Reference:

The XtremIO Performance Documents discuss different workload testing parameters, including I/O

size and limits, which are relevant to sizing the array1.

The Dell XtremIO Design Achievement document outlines the critical components and best practices

for designing solutions with XtremIO and X2 storage systems, which would include considerations for

workload sizing2.

=========================

For ESXi disks provisioned by XtremIO, the recommended disk format is ‘Eager Zeroed Thick’. This

format is preferred because it allocates all the space required for the disk at the time of creation and

zeroes out any previous data on the physical media. Compared to other formats, ‘Eager Zeroed Thick’

provides better performance due to the pre-zeroing of the disk space, which helps in avoiding any

latency that might occur during write operations as there is no need to zero out blocks before writing

new data.

Reference:

While the specific Dell XtremIO Design document was not available, the general best practice for

high-performance storage solutions like XtremIO is to use ‘Eager Zeroed Thick’ provisioning for VMs

running on VMware ESXi to ensure optimal performance1.

Additional information on disk provisioning and best practices can be found in VMware’s official

documentation and Dell EMC’s storage configuration guides2.

=========================

A response code of 400 in the context of RESTful API operations generally indicates that the request

made by the client was malformed or contains a syntax error. This means that the data stream sent

by the client to the server did not follow the rules as defined by the API specification. It could be due

to various reasons such as incorrect request format, missing required parameters, or invalid

parameter values1.

In the case of the Dell XtremIO environment, if a customer receives a 400 response code when

attempting to create a volume via the RESTful API, it is likely due to a syntax error in the API

command. The customer should review the API request to ensure that it conforms to the expected

format and includes all necessary information as per the XtremIO API documentation2.

Reference:

Community discussions on the Dell Technologies website where similar issues have been addressed

and the cause of the 400 error was identified as a syntax error in the RESTful API command2.

General information about the meaning of a 400 Bad Request error in REST APIs, which supports the

conclusion that a syntax error is the cause1.

=========================