Volume 15, Issue IV
Hello and welcome to this April 2015 Server and StorageIO update newsletter.
This months newsletter has a focus on cloud and object storage for bulk data, unstructured data, big data, archiving among other scenarios.
Enjoy this edition of the Server and StorageIO update newsletter and watch for new tips, articles, StorageIO lab report reviews, blog posts, videos and Podcasts along with in the news commentary appearing soon.
April StorageIOblog posts include:
View other recent as well as past blog posts here
April Newsletter Feature Theme
Cloud and Object Storage Fundamentals
There are many facets to object storage including technology implementation, products, services, access and architectures for various applications and use scenarios. The following is a short synopsis of some basic terms and concepts associated with cloud and object storage.
Common cloud and object storage terms
Account or project – Top of the hierarchy that represent owner or billing information for a service that where buckets are also attached.
Availability Zone (AZ) can be rack of servers and storage or data center where data is spread across for storage and durability.
Example of some AWS Regions and AZ’s
Bucket or Container – Where objects or sub-folders containing objects are attached and accessed. Note in some environments such as AWS S3 you can have sub-folders in a bucket.
Connector or how your applications access the cloud or object storage such as via an API, S3, Swift, Rest, CDMI, Torrent, JSON, NAS file, block of other access gateway or software.
Durability – Data dispersed with copies in multiple locations to survive failure of storage or server hardware, software, zone or even region. Availability = Access + Durability.
End-point – Where or what your software, application or tool and utilities or gateways attach to for accessing buckets and objects.
Ephemeral – Temporary or non-persistent
Eventual consistency – Data is eventually made consistency, think in terms of asynchronous or deferred writes where there is a time lag vs. synchronous or real-time updates.
Immutable – Persistent, non-altered or write once read many copy of data. Objects generally are not updated, rather new objects created.
Via Cloud Virtual Data Storage (CRC)
Object – Byte (or bit) stream that can be as small as one byte to as large as several TBytes (some solutions and services support up to 5TByte sized objects). The object contains what ever data in any organization along with meta data. Different solutions and services support from a couple hundred KBytes of meta-data to MBytes worth of meta-data. In terms of what can be stored in an object, anything from files, videos, images, virtual disks (VMDK’s, VHDX), ZIP or tar files, backup and archive save sets, executable images or ISO’s, anything you want.
OPS – Objects per second or how many objects accessed similar to a IOP. Access includes gets, puts, list, head, deletes for a CRUD interface e.g. Created, Read, Update, Delete.
Region – Location where data is stored that can include one or more data centers also known as Availability Zones.
Sub-folder – While object storage can be accessed in a flat name space for commonality and organization some solutions and service support the notion of sub-folder that resemble traditional directory hierarchy.
Learn more in Cloud Virtual Storage Networking (CRC) and www.objectstoragecenter.com
OpenStack Manila (e.g. Folders and Files)
AWS recently announced their new cloud based Elastic File Storage (EFS) to compliment their existing Elastic Block Storage (EBS) offerings. However are you aware of what is going on with cloud files within OpenStack?
For those who are familiar with OpenStack or simply talk about it and Swift object storage, or perhaps Cinder block storage, are you aware that there is also a file (NAS or Network Attached Storage) component called Manila?
In concept Manila should provide a similar capability to what AWS has recently announce with their Elastic File Service (EFS), or depending on your perspective, perhaps the other way around. If you are familiar and have done anything with Manila what are your initial thoughts and perspectives.
What this all means
People routinely tell me this is the most exciting and interesting times ever in servers, storage, I/O networking, hardware, software, backup or data protection, performance, cloud and virtual or take your pick too which I would not disagree.
However, for the past several years (no, make that decade), there is new and more interesting things including in adjacent areas.
I predict that at least for the next few years (no, make that decades), we will continue to see plenty of new and interesting things, questions include.
However, what’s applicable to you and your environment vs. simply fun and interesting to watch?
Ok, nuff said, for now