Recap of my session at Microsoft TechEd 2014 – Microsoft Azure Storage
Azure Storage is a cloud based storage offering which is hosted in 16 regions across the world.
Links:
blogs.msdn.com/b/windowsazurestorage
1> BLOBS
- An object store for unstructured data
- REST API access only
- Flat namespace
-
Up to 60 MB/sec throughput
-
Up to 1TB/BLOB
-
Charged by storage actual consumption
2> Disks
- Network mounted virtual disk (inside of Azure for Azure-based VMs)
- Stored in BLOB
- up to 1TB Disk
- Max 8KB IOPS – 500 IOPS
- Snapshot and copy
- Up to 60MB/sec throughput
- Disks are exclusive to a single VM
3> Tables
- noSQL cloud store
- Odata protocol (AtomPub or JSON)
4> Queues
- Reliable messaging system
- Low latency, high throughput
5> Files (new offering as of May 2014)
- Azure File Share – SMB and REST
- SMB 2.1
- Up to 5TB File Shares
- Up to 1TB file
- SMB 2.1, access only available within region, otherwise use REST
- REST accessible from within Azure and from Internet
- True directory structure
- Up to 60 MB/sec throughput per share
- Charged based file size
- Can’t boot VM’s off of Azure File
- Max 8KB IOPS – 1000 IOPS
Azure Storage Protection options:
LRS – Local Redundant Storage –
3 copies within a zone (facility)
ZRS – Zone Redundant Storage (for block BLOBs only)
3 copies between zones but within the same Region
GRS – 6 copies of data across two regions (3 in each region) – Async consistent
Read-only access to GRS
Read-only access to secondary data even if primary in unavailable
Strongly consistent within the primary region
Eventually consistent across regions
All writes go to primary
Azure import/Export
Move Tbs of data into and out of Azure BLOBS by shipping disks
Encrypted with BitLocker
Azure Copy for files
Utility to copy files into Azure Files (via REST API)
Tools and libraries:
Client libraries
.NET, Java, C++
Powershell commands
CLI tools
What’s coming:
iOS library
Javascript
Zone redundant storage for BLOBs
Reblogged this on atalupadhyay.