A summary of the facts and figures for Microsoft Azure Services.
All figures have been taken from official Microsoft web sites (see the refrences at the bottom).
This reference does not attempt to cover cost. Please refer to the Azure Pricing Calculator instead.
This guide is incomplete. Please fork, adjust and send a pull request (please include a link to official Microsoft documentation for any information added to the page).
This reference is available as a single page or split into multiple pages.
- Asia Pacific
- Asia Pacific East (Hong Kong)
- Asia Pacific Southeast (Singapore)
- Europe
- Europe North (Ireland)
- Europe West (Netherlands)
- United Kingdom (TBA)
- United States
- US North Central (Illinois)
- US South Central (Texas)
- US East (Virginia)
- US East 2 (Virginia)
- US West (California)
- US Central (Iowa)
- Japan
- Japan East (Tokyo)
- Japan West (Osaka)
- China
- Beijing
- Shanghai
- Australia
- Austalia East (Sydney)
- Australia Southeast (Melbourne)
- South America
- Brazil South (São Paulo)
- India *Mumbai *Chennai *Pune
*Note: Microsoft Azure services in China Data Centres are only available for use from within China itself. These data centers are disconnected from rest of the Microsoft Azure Data Centres. You can visit the chinese version of Microsoft Azure at: http://windowsazure.cn/
To find your nearest Data Center, use the Azure Speed Test.
View the locations on a map (uses a geojson gist).
A list of services available in each Data Center is available here.
- Trust Center - Location of Customer Data
- Microsoft becomes the first global public cloud provider in China with the general availability of Microsoft Azure in the region
- Azure Regions
Virtual Machine sizes:
| Size | Cores | RAM | Disk [a] (Web/Worker) |
Bandwidth (indicative) |
|---|---|---|---|---|
| X-Small | Shared | 768 MB | 19 GB | 5 Mbps |
| Small | 1 | 1.75 GB | 224 GB | 100 Mbps |
| Medium | 2 | 3.5 GB | 489 GB | 200 Mbps |
| Large | 4 | 7 GB | 999 GB | 400 Mbps |
| X-Large | 8 | 14 GB | 2,039 GB | 800 Mbps |
| A5 | 2 | 14 GB | 489 GB | 200 Mbps |
| A6 | 4 | 28 GB | 999 GB | 400 Mbps |
| A7 | 8 | 56 GB | 2,039 GB | 800 Mbps |
[a] Disk (Web/Worker) is the local storage available on Cloud Services
- Max disk performance: 500 IOPS per disk
- Max roles per deployment: 25
- Max endpoints per deployment: 25 input endpoints, 25 internal endpoints
- SLA: 99.95% (with two or more role instances in different fault and upgrade domains)
| Size | CPU | RAM | Disk |
|---|---|---|---|
| A8 | 8 @ 2.6 GHz(Intel® Xeon® E5-2670) | 56 GB (DDR3-1600 MHz) | 1,770 GB |
| A9 | 16@ 2.6 GHz(Intel® Xeon® E5-2670) | 112 GB (DDR3-1600 MHz) | 1,770 GB |
Notes
- PaaS only
- 10 Gbps Ethernet
- 40 Gbps back end, Remote Direct Memory Access (RDMA) capable.
- RDMA is only enabled through applications that use the Network Direct interface. (i.e. HPC Pack 2012 R2 on Windows Server 2012 or Windows Server 2008 R2. NOT supported on Windows Server 2012 R2)
- Only available in West Europe. (Feb 2014)
- Cannot be deployed by using a cloud service that is part of an existing affinity group. Likewise, an affinity group with a cloud service containing A8 or A9 instances cannot be used for deployments of other instance sizes.
- Number of Roles Allowed In Windows Azure Deployments Increased To 25
- Windows Azure, Increase Cores count
- Virtual Machine and Cloud Service Sizes for Azure
- New High Performance Capabilities for Windows Azure
- A8 and A9 Compute Intensive Instances: Quick Start with HPC Pack
- Virtual Machine and Cloud Service Sizes for Azure
Virtual Machine sizes:
| Size | Cores | RAM | Disk [a] (VM) |
Bandwidth (indicative) |
|---|---|---|---|---|
| X-Small | Shared | 768 MB | 20 GB | 5 Mbps |
| Small | 1 | 1.75 GB | 70 GB | 100 Mbps |
| Medium | 2 | 3.5 GB | 135 GB | 200 Mbps |
| Large | 4 | 7 GB | 285 GB | 400 Mbps |
| X-Large | 8 | 14 GB | 605 GB | 800 Mbps |
| A5 | 2 | 14 GB | 135 GB | 200 Mbps |
| A6 | 4 | 28 GB | 285 GB | 400 Mbps |
| A7 | 8 | 56 GB | 605 GB | 800 Mbps |
[a] Disk (VM) is the temp drive available on Virtual Machines
- Max disk performance: 500 IOPS per disk
- Max size of Virtual Machine OS Disk: 127 GB
- Max roles per deployment: 25
- Max endpoints per deployment: 25 input endpoints, 25 internal endpoints
- SLA: 99.95% (with two or more role instances in different fault and upgrade domains)
- Virtual Machine and Cloud Service Sizes for Azure
- New High Performance Capabilities for Windows Azure
- Virtual Machine and Cloud Service Sizes for Azure
- Windows Azure, Increase Cores count
- Max free web sites per Data Center: 10
- Max shared websites per Data Center: 100
- Max standard websites per Data Center: 500
- Free tier limitations: 1% CPU per hour, 2.5 minutes of CPU every 5 minutes, 165 MB data out per day, 1024 MB storage, 1024 MB RAM per hour
- Max instance count in standard mode: 10 (large instances)
- SLA: 99.9%
- Azure: ExpressRoute Dedicated Networking, Web Site Backup Restore, Mobile Services .NET support, Hadoop 2.2, and more
- Azure Websites
- How to Scale Web Sites
- United States
- East: Atlanta, Miami, New York, Washington D.C.
- North Central: Chicago
- South Central: Dallas
- West: Los Angeles, San Jose, Seattle
- Europe
- North: Copenhagen, Helsinki, Stockholm, Vienna, Warsaw
- West: Amsterdam, Frankfurt, Milan, London, Madrid, Paris
- Asia-Pacific/Rest of World
- Asia Pacific East: Batam, Hong Kong, Jakarta, Kaohsiung, Singapore
- Asia Pacific Southeast: Melbourne, Sydney
- Japan East: Tokyo
- Japan West: Osaka
View the locations on a map (uses a geojson gist).
Azure Content Delivery Network (CDN) Node Locations
Free Tier
- Max number of indexes: 3
- Max document count: 10,000
- Max storage size: 50 MB
- Max partitions: N/A
- Max replicas: N/A
- Max search units: N/A
Standard Tier
- Max number of indexes: 50 per Search service
- Max document count: 15 million per partition
- Max storage size: 25 GB per partition
- Max partitions: 12 per Search service
- Max replicas: 6 per Search service
- Max search units: 36 per Search service
API Key Limits
- Max admin keys per service: 2
- Max query keys per service 50
Request Sizes
- Max size of request : 16 MB
- Max URL length: 8 KB
- Max documents per batch: 1000 documents per batch of index uploads/merges/deletes
- Characters of text for Suggest API: Min 3, Max 25 characters
- Max fields in $orderby clause: 32
Response Sizes
- Max documents returned per page of search results: 1000
- Maximum suggestions returned per Suggest API request: 10
http://msdn.microsoft.com/en-us/library/azure/dn798934.aspx
- Max number of service bus namespaces: 50 per Azure Acccount
- Expected transactions per second: 2,000 per queue
- Expected latency: 100 ms
- Max message size: 256 KB (Max header size 64kb)
- Max message TTL: Unlimited
- Max queue size: up to 5 GB
- Queue size options: 1-5 GB (queue cannot be resized once created)
- Max number of queues: 10,000 per service bus namespace
soft limit - Ordering guarantee: Yes (through the use of messaging sessions)
- Lease / Lock duration: 60 seconds (default) (can bre renewed using the RenewLock API)
- Naming constraints: queues names no longer than 260 chars (case sensitive)
- SLA: 99.9%
- Queue Space Consumption: Number Of Messages * (Message Body Size + Message Header Size)
- Topic Space Consumption: Number Of Messages * (Message Body Size + (Message Header Size * Number Of Subscriptions))
Capacity Approximations:
| Message Size | 1 GB Queue | 2 GB Queue | 3 GB Queue | 4 GB Queue | 5 GB Queue |
|---|---|---|---|---|---|
| 1 KB | 1,041,790 | 2,059,920 | 3,128,550 | 4,186,400 | 5,238,750 |
| 10 KB | 102,996 | 208,358 | 312,537 | 416,716 | 520,895 |
| 50 KB | 20,857 | 41,792 | 62,507 | 83,343 | 104,179 |
| 100 KB | 10,466 | 20,836 | 31,254 | 41,672 | 52,090 |
| 250 KB | 4,191 | 8,334 | 12,501 | 16,669 | 20,836 |
- SLA: 99.9%
- Windows Azure Queues and Windows Azure Service Bus Queues - Compared and Contrasted
- Capacity Planning for Service Bus Queues and Topics
- Windows Azure Service Bus Quotas
Basic Tier
- Max size: 2 GB
- Self Service Restore window: 7 days
- Geo replication: Active
- Performance level: Basic
Standard Tier
- Max size: 250 GB
- Self Service Restore window: 35 days
- Geo replication: Active
- Performance level: S0 S1 S2 S3
Premium Tier
- Max size: 500 GB (P1 - P6) or 1024 GB (P11 and P15 only)
- Self Service Restore window: 35 days
- Geo replication: max 4 readable secondaries
- Performance level: P1 P2 P4 P6 P11 P15
All Tiers
- SLA: 99.99%
- Max database servers: 6 per Azure Account
soft limit - Max databases per server: 150 (including master)
soft limit - Max number of database backup schedules: 10 per subscription
Performance Levels
| Service Tier/Performance Level | DTU [1] | Max Worker Threads | Max Sessions | Benchmark Transaction Rate [2] | Predictability [3] |
|---|---|---|---|---|---|
| Basic | 5 | 20 | 100 | 3,467 transactions/hour | Good |
| Standard/S0 | 10 | 50 | 200 | 283 transactions/minute | Better |
| Standard/S1 | 20 | 50 | 200 | 283 transactions/minute | Better |
| Standard/S2 | 50 | 100 | 500 | 1,470 transactions/minute | Better |
| Standard/S3 | 100 | - | - | - | Better |
| Premium/P1 | 125 | 200 | 2,000 | 98 transactions/second | Best |
| Premium/P2 | 250 | 400 | 4,000 | 192 transactions/second | Best |
| Premium/P4 | 500 | 1,600 | 16,000 | 730 transactions/second | Best |
| Premium/P6 | 1000 | - | - | - | Best |
| Premium/P11 | 1750 | - | - | - | Best |
| Premium/P15 | 4000 | - | - | - | Best |
[1] Database Throughput Unit (DTU): The resources powering each performance level are represented in DTUs. It combines CPU, memory, physical reads, and transaction log writes into a single unit. A performance level with 5 DTUs has five times more power than a performance level with 1 DTU.
[2] Transaction Rate: Transaction rate is the metric produced by the Azure SQL Database Benchmark. It is reported in transactions per unit-of-time, counting all transaction types. In the Azure SQL Database Benchmark, throughput is reported in transaction per unit-of-time, with the time scale being shorter for higher classes of service, as shown below.
[3] Predictability: Consistency of response time is an indication of performance predictability. A database which achieves a more stringent response time constraint delivers more predictable performance. In the Azure SQL Database Benchmark, the response time constraint varies with class of service, with higher classes of service having to satisfy a more stringent response time requirement.
- Windows Azure Table Storage and Windows Azure SQL Database - Compared and Contrasted
- Scott Gu's Blog - Azure: 99.95% SQL Database SLA, 500 GB DB Size, Improved Performance Self-Service Restore, and Business Continuity
- Azure SQL Database introduces new service tiers
- Azure SQL Database Service Tiers and Performance Levels
- Azure SQL Database Business Continuity
- Total capacity: 500 TB
- Expected transactions per second: 20,000 per account
- Total bandwidth for US Regions: 20 GB in 30 GB out per second (10 GB and 20 GB respectively for geo redundant)
- Total bandwidth for Europeans and Asian Regions: 10 GB in 15 GB out per second (5 GB and 10 GB respectively for geo redundant)
- Maximum number of accounts: 100
- SLA: 99.9%
- Expected transactions per second: 500 per blob
- Expected bandwidth: 60 MB per second
- Max blob size: 200 GB (block blob) 1 TB (page blob)
- Expected transactions per second: 2,000 per table partition
- Max table size: 200 TB
- Max row size: 1 MB
- Maximum key size 1 KB for PartitionKey and RowKey
- Max value size: 64 KB
- Max number of values per row: 255 (including partition key, row key and timestamp)
- Max results per query: 1,000
- Max batch insert size: 4 MB (must have the same partition key)
- Expected transactions per second: 2,000 per queue
- Expected latency: 10 ms
- Max message size: 64 KB
- Max message TTL: 7 days
- Max queue size: 200 TB
- Ordering guarantee: NO
- Lease / Lock duration: 30 seconds (default) 7 days(max)
- Max number of queues: unlimited
- Naming constraints: queues names no longer than 63 chars (lower case)
- Number of messages you can pop in one transaction: 32
- Max share size: 5 TB per share
- Expected bandwidth: 60 MB per second per share
- Max file size: 1 TB per file
- Max 8 KB IOps: 1000 IOps
Note only accessible over SMB 2.1 protocol from a VM within the same region. REST access from anywhere. Read-only geo redundancy not available.
- Disable Nagle for small messages (<1.4k)
- Disable Expect 100-Continue*
- Increase default connection limit (ex. 100+)
- Take advantage of .Net 4.5 GC
ServicePointManager.UseNagleAlgorithm = false;
ServicePointManager.Expect100Continue = false;
ServicePointManager.DefaultConnectionLimit = 100;- Windows Azure Queues and Windows Azure Service Bus Queues - Compared and Contrasted
- Windows Azure Table Storage and Windows Azure SQL Database - Compared and Contrasted
- Storage
- Understanding the Table Service Data Model
- MVA - Design & Implementation Jump Start
- Azure Storage Scalability and Performance Targets
- Introducing Microsoft Azure File Service
- Azure Subscription and Service Limits, Quotas, and Constraints
- Naming constraints: 253 characters in total for the *.trafficmanager.net domain, maximum of 38 sub domains, each of which can be up to 63 characters
- TTL recommendation: 5 minutes
- Application polling: 30 second frequency, 10 second timeout, 3 retries before marked unavailable
- SLA: 99.99%
- Max VPN client endpoints: 250
- SLA: 99.9%
Co-admins per subscription: 200
| SLA % | Downtime per day | Downtime per week | Downtime per month | Downtime per year |
|---|---|---|---|---|
| 99.9 | 1 minute 26 seconds | 10 minutes 5 seconds | 43 minutes 50 seconds | 8 hours 46 minutes |
| 99.95 | 43 seconds | 5 minutes 2 seconds | 21 minutes 55 seconds | 4 hours 23 minutes |
| 99.99 | 8 seconds | 1 minute | 4 minutes 23 seconds | 52 minutes 23 seconds |
| 99.999 | 1 second | 6 seconds | 26 seconds | 5 minutes 15 seconds |
Thanks to the contributors:
...and thanks to Rob Blackwell for the idea.