Month: January 2021
Shared Storage Options in Azure: Part 3 – Azure Storage Services
Reading Time: 8 minutesWelcome to Part 3 of this 5-part Series on Shared Storage Options in Azure. In this post I’ll be covering Azure Storage Services. You may be thinking to yourself, wait a minute, what have we been talking about this whole time then? Azure Storage Services is easiest thought of being the term used for the services offered under an Azure Storage Account (Blob, File, Queue, Table). Given the context of this series, I’ll be discussing Azure Blob Storage and Azure File storage in this post. Though, I do want to add a disclaimer that technically Queue, and Table Storage can be “shared” also since multiple apps can call the same Queue or Table using the APIs. Since the focus here is more on the system-level, I’m not going to cover those two, but I’ll add some links to documentation where you can read more.
- Part 1: Azure Shared Disks
- Part 2: IaaS Storage Server
- Part 3: Azure Storage Services
- Part 4: Azure NetApp Files (coming soon)
- Part 5: Conclusion (coming soon)
Azure Blob Storage:
In the majority of cases, when people discuss “cloud storage” they’re talking about Blob – binary large object. What this service allows us to do is store massive amounts of unstructured “objects” in Azure. There are a couple ways we can Blob storage as shared storage from a system-level.
Shared Blob Storage:
As I mentioned in the introduction, all Azure Storage Services can be accessed over HTTP/S via API or using any of the client libraries. This means that they can all technically be “shared” storage, but what about system-level access? While I find most applications and solutions can be adapted using a client library, there is a project called “Blobfuse” which can be used for more traditional applications.
Blobfuse is an open-source project on GitHub which uses the libfuse library to pair together the Linux FUSE kernel module and the Azure Blob REST APIs to create a virtual filesystem. The result of this configuration is a mount point on a Linux machine directly to a Blob Storage Account. There can be certain challenges in using Blobfuse though, for example the result is NOT a POSIX-compliant filesystem and if you use mount the same Blob Storage from multiple machines you should keep those limitations in mind.
The default configuration for the setup of Blobfuse is to have your Storage Account name and Access Key in a plain-text configuration file sitting on your server, which is not ideal from a security perspective and should be noted. However, it is possible to use a Managed Service Identity with Blobfuse which significantly improves the security posture of the deployment (if you use a System Assigned Managed Identity) and something I would recommend over the default configuration. Lastly, Blobfuse is not available on Windows – Linux only.
As of the time of writing this blog post (January, 2021) NFS is not yet Generally Available (GA) on Blob Storage, but NFS 3.0 has been in preview since July 2020 . Once this goes GA I will update this post with that information, but won’t quote this as an option until that point.
Lastly, from a backup and disaster recovery perspective, Azure Blob Storage supports snapshots as well as Point-in-Time restore for block blobs.
Typical Use Cases:
The majority of the use cases I’ve seen that use Blob as shared storage at the system level are wanting to use consumption-based cloud storage without the overhead or limitations of a managed disk. Specifically, in applications that don’t support SMB natively and require a local mount point, that’s where Blobfuse comes into play. I have seen this with a lot of apps that are migrated into the cloud and want lower cost, higher capacity than is available from managed disks with more legacy applications where this may be the case. I’ve also seen this configuration with many HPC applications since NFS as an access protocol is not yet GA for Blob storage.
Cost, Performance, Availability and Limitations:
The cost of using Blob storage is always the same regardless of the access protocol, since as of now it all ends up going through the Azure Storage API anyways.
Blob storage is incredibly performant. There are two tiers of Blob storage, Standard and Premium. In most cases, Standard will be the appropriate tier. Premium is for storage that needs single-digit transaction times and is better suited for larger block sizes (256KiB+). Though do keep in mind, that similar to my comparison of Managed Disk Types and the cost calculation of capacity and transaction costs, in some scenarios Premium Block Blob Storage may be cheaper.
If you’re using a standard Blob storage account (not configured with a Hierarchical Namespace) which is most common, you’ll enjoy the following performance (as of January, 2021).
Image Reference: https://docs.microsoft.com/en-us/azure/azure-resource-manager/management/azure-subscription-service-limits#storage-limits
When configuring containers in your Blob Storage Account you’ll notice an access tier setting with the options “Hot, Cool, Archive”. I covered Archive Storage a few years back but it’s not really relevant to this topic. What is relevant though is the different between Hot and Cool storage. There seems to still be a lot of confusion around the difference between the two, but at it’s core the main different is transaction cost.
Similar to the difference between Premium and Standard SSD Managed disks, Hot Blob storage has a higher capacity cost but a lower transaction cost while Cool Blob Storage has a lower capacity cost and a higher transaction cost. If you’re storing data that is infrequently accessed but still needs to be constantly available, Cool Blob Storage is the way to go. If you’re storing data that has a lot of transactions then Hot Blob storage is your best bet. Don’t get caught up in the “per GB” sticker on each tier – this can be misleading to the resulting cost depending on your workload characteristics.
As far as durability and availability goes, Storage Accounts have a few different options depending on the storage service being used: LRS, ZRS, GRS, ZGRS, GRS-RA. There is a lot of information on these different redundancy levels, so take a look at the durability and availability table below and if you want to read more, click the link below.
Image Reference: https://docs.microsoft.com/en-us/azure/storage/common/storage-redundancy
Additional Reading on Understanding Azure Storage Redundancy Offerings: https://techcommunity.microsoft.com/t5/azure-storage/understanding-azure-storage-redundancy-offerings/ba-p/1431700
Lastly, I recently re-created an outdated version of an infographic on capacity limits for Azure Storage Accounts and I thought I would share that here.
Feel free to reference this image using the following: https://urls.hansencloud.com/azure-storage-limits
Azure Files:
Azure Files is another storage service under the Azure Storage Account and has similar shared features, but some very distinct to itself as well. The primary purpose of Azure Files is to provide file-level storage services like you would get from a Network Attached Storage appliance or a Server providing those access protocols to a filesystem share. Azure Files provides SMB access (as well as HTTP/S API) to provisioned shares.
Access Methods:
Like I mentioned, the primary access methods for Azure Files are through the API or by using SMB (NFS v4.1 is currently in preview so I won’t be considering it an option in this post as of right now, but will update it when it goes GA). Even though SMB is most typically used in with Windows machines, shares on Azure Files can be used by Windows, Linux, or even MacOS.
Something really interesting about Azure Files though is its Azure File Sync capability which allows for a centralized file share in Azure Files which can be facilitated through agents deployed on Windows Servers which then act as a cache for the Azure Files data. This is particularly interesting because it allows the Server itself to present whichever access method it would like to the client, but use the backing of a centralized Azure Files Share.
The way Azure File Sync works at a high-level is a File Share is created, then linked to what is called a “sync group”, which facilitates the registrations from any agents deployed on Windows Servers (in Azure or on-prem).
Azure Files also allows, in conjunction with typical access key authentication, Active Directory-based authentication options . The ability to use this type of AuthN directly on the Azure Files PaaS endpoint is really interesting and makes it a great choice for a solution where you want to leverage the identity systems you already have in place. It’s also worth noting that if you’re using Azure File Sync, the deployed agent is the only one communicating with the File Share directly and the access to the data locally can be controlled through whichever method you prefer (SMB ACLs with ADDS, for example).
Lastly, from a backup and disaster recovery prospective, Azure Files supports snapshots in addition to native integration with Azure Backup.
Typical Use Cases:
I see a mix of uses with Azure Files. A mix between using it for a file-based backend for various applications and services to an environment where the data is access directly by users. A scenario I’ve run into more frequently though is when companies want to replace traditional on-prem File Servers and even things like DFS. Anywhere you want to leverage SMB in a fully managed PaaS way, Azure Files is for you.
Cost, Performance, Availability and Limitations:
Similar to Blob Storage, Azure Files has multiple Tiers to help optimize for performance and Cost.
Image Reference: https://azure.microsoft.com/en-us/pricing/details/storage/files/
Again, these tiers are priced based on capacity (provisioned or consumed) in combination with transactions and any snapshots or backups.
Performance for Azure Files is based on whether or not you use a standard storage account or a specific “Azure Files” storage account SKU which will enable “Premium” File Shares. The performance specifications for a standard storage account (eg. General Purpose v2) are the same as the limits posted for the blob storage earlier. If you’re using Premium Files though, here are the performance targets.
Image Reference: https://docs.microsoft.com/en-us/azure/storage/files/storage-files-scale-targets
Keep in mind that the 100TB limit is per share, and you can create multiple (just like you would with traditional file shares) up to the limit of the Storage Account (5 PB by default, as stated earlier in this post).
Lastly, availability for Azure Files is no different than Blob since they’re both contained by a storage account and will be subject to the availability and durability of the storage account data redundancy setting.
Pros and Cons:
Okay, here we go with the Pros and Cons for using an Azure Storage Services (Blob & File) for your shared storage configuration on Azure.
Pros:
- Both are PaaS and fully managed which greatly reduces operational overhead.
- Significantly higher capacity limits as compared to IaaS.
- Ability to migrate workloads as-is and use existing storage configurations when using SMB or Blobfuse.
- Ability to use Active Directory Authentication in Azure Files
- Both integrate with native backup solutions.
- Both integrate with Azure Defender for Storage.
Cons:
- Blobfuse stores connection information in plain-text, by default.
- Does not support older access protocols like iSCSI.
Alright, that’s it for Part 3 of this blog series – Shared Storage on Azure Storage Services. Please reach out to me in the comments, on LinkedIn, or Twitter with any questions or comments about this post, the series, or anything else!
- Part 1: Azure Shared Disks
- Part 2: IaaS Storage Server
- Part 3: Azure Storage Services
- Part 4: Azure NetApp Files (coming soon)
- Part 5: Conclusion (coming soon)
Shared Storage Options in Azure: Part 2 – IaaS Storage Server
Reading Time: 9 minutesRecently, I posted the “Shared Storage Options in Azure: Part 1 – Azure Shared Disks” blog post, the first in the 5-part series. Today I’m posting Part 2 – IaaS Storage Server. While this post will be fairly rudimentary insofar as Azure technical complexity, this is most certainly an option when considering shared storage options in Azure and one that is still fairly common with a number of configuration options. In this scenario, we will be looking at using a dedicated Virtual Machine to provide shared storage through various methods. As I write subsequent posts in this series, I will update this post with the links to each of them.
- Part 1: Azure Shared Disks
- Part 2: IaaS Storage Server
- Part 3: Azure Storage Services
- Part 4: Azure NetApp Files (coming soon)
- Part 5: Conclusion (coming soon)
Virtual Machine Configuration Options:
Compute:
While it may not seem vitally important, the VM SKU you choose can impact your ability to provide storage capabilities in areas such as Disk Type, Capacity, IOPS, or Network Throughput. You can view the list of VM SKUs available on Azure at this link. As an example, I’ve clicked into the General Purpose, Dv3/Dvs3 series and you can see there are two tables that show upper limits of the SKUs in that family.
In the limits for each VM you can see there are differences between Max Cached and Temp Storage Throughput, Max Burst Cached and Temp Storage Throughput, Max uncached Disk Throughput, and Max Burst uncached Disk Throughput. All of these represent very different I/O patterns, so make sure to look carefully at the numbers.
Below are a few links to read more on disk caching and bursting:
- Disk Caching: https://docs.microsoft.com/en-us/azure/virtual-machines/premium-storage-performance#disk-caching
- Disk Bursting: Managed disk bursting – Azure Virtual Machines | Microsoft Docs
You’ll notice when you look at VM SKUs that there is an L-Series which is “storage optimized”. This may not always be the best fit for your workload, but it does have some amazing capabilities. The outstanding feature of the L-Series VMs are the locally mapped NVMe drives which as of the time of writing this post on the L80s_v2 SKU can offer 19.2TB of storage at 3.8 Million IOPS / 20,000 MBPS.
The benefits of these VMs are the extremely low latency, and high throughput local storage, but the caveat to that specific NVMe storage is that it is ephemeral. Data on those disks does not persist a reboot. This means it’s incredibly good at serving from a local cache, tempdb files, etc. though its not storage that you can use for things like a File Server backend (without some fancy start-up scripts, please don’t do this…). You will note that the maximum uncached throughput is 80,000 IOPS / 2,000 MBPS for the VM, which is the same as all of the other high spec VMs. As I am writing this, no Azure VM allows for more than that for uncached throughput – this includes Ultra Disks (more on that later).
For more information on the LSv2 series, you can read more here: Lsv2-series – Azure Virtual Machines | Microsoft Docs
Additional Links on Azure VM Storage Design:
- Azure Premium Storage: Design for high performance – Azure Virtual Machines | Microsoft Docs
- Virtual machine and disk performance – Linux – Azure Virtual Machines | Microsoft Docs
Networking:
Networking capabilities of the Virtual Machine are also important design decisions when considering shared storage, both in total throughput and latency. You’ll notice in the VM SKU charts I posted above when talking about performance there are two sections for networking, Max NICs and Expected network bandwidth Mbps. It’s important to know that these are VM SKU limitations, which may influence your design.
Expected network bandwidth is pretty straight forward, but I want to clarify that the number of Network Interfaces you mount to a VM does not change this number. For example, if your expected network bandwidth is 3200 Mbps and you have an SMB share running on that single NIC, adding a second NIC and using SMB multi-channel WILL NOT increase the total bandwidth for the VM. In that case you could expect each NIC to potentially run at 1,600 Mbps.
The last networking feature to take into consideration is Accelerated Networking. This feature allows for SR-IOV (Single Root I/O Virtualization), which by bypassing the host CPU and offloading the network traffic directly to the Network Interface can dramatically increase performance by reducing latency, jitter, and CPU utilization.
Image Reference: Create an Azure VM with Accelerated Networking using Azure CLI | Microsoft Docs
Accelerated Networking is not available on every VM though, which makes it an important design decision. It’s available on most General Purpose VMs now, but make sure to check the list of supported instance types. If you’re running a Linux VM, you’ll also need to make sure it’s a supported distribution for Accelerated Networking.
Storage:
In an obvious step, the next design decision is the storage that you attach to your VM. There are two major decision types when selecting disks for you VM – disk type, and disk size.
Disk Types:
Image Reference: https://docs.microsoft.com/en-us/azure/virtual-machines/disks-types
As the table above shows, there are three types of Managed Disks (https://docs.microsoft.com/en-us/azure/virtual-machines/managed-disks-overview ) in Azure. At the time of writing this, Premium/Standard SSD and Standard HDD all have a limit of 32TB per disk. The performance characteristics are very different, but I also want to point out the difference in the pricing model because I see folks make this mistake very often.
Disk Type: | Capacity Cost: | Transaction Cost: |
Standard HDD | Low | Low |
Standard SSD | Medium | Medium |
Premium SSD | High | None |
Ultra SSD | Highest (Capacity/Throughput) | None |
Transaction costs can be important on a machine whose sole purpose is to function as a storage server. Make sure you look into this before a passing glance shows the price of a Standard SSD lower than a Premium SSD. For example, here is the Azure Calculator output of a 1 TB disk across all four types that averages 10 IOPS * ((10*60*60*24*30)/10,000) = 2,592 transaction units.
Sample Standard Disk Pricing:
Sample Standard SSD Pricing:
Sample Premium SSD Pricing:
Sample Ultra Disk Pricing:
The above example is just an example, but you get the idea. Pricing gets strange around Ultra Disk due to the ability to configure performance (more on that later). Though there is a calculable break-even point for disks that have transaction costs versus those that have a higher provisioned cost.
For example, if you run an E30 (1024 GB) Standard SSD at full throttle (500 IOPS) the monthly cost will be ~$336, compared to ~$135 for a P30 (1024 GB) Premium SSD, with which you get x10 the performance. The second design decision is disk capacity. While this seems like a no-brainer (provision the capacity needed, right?) it’s important to remember that with Managed Disks in Azure, the performance scales with, and is tied to, the capacity of the disk.
Image Reference: https://docs.microsoft.com/en-us/azure/virtual-machines/disks-types#disk-size-1
You’ll note in the above image the Disk Size scales proportionally with both the Provisioned IOPS and Provisioned Throughput. This is to say that if you need more performance out of your disk, you scale it up and add capacity.
The last note on capacity is this, if you need more than 32TB of storage on a single VM, you simply add another disk and use your mechanism for combining that storage (Storage Spaces, RAID, etc.). This same method can be used to further tweak your total IOPS, but make sure you take into consideration the cost of each disk, capacity, and performance before doing this – most often it’s an insignificant cost to simply scale-up to the next size disk. Last but not least, I want to briefly talk about Ultra Disks – these things are amazing!
Unlike with the other disk types, this configuration allows you to select your disk size and performance (IOPS AND Throughput) independently! I recently worked on a design where the customer needed 60,000 IOPS, but only needed a few TB of capacity, this is the perfect scenario for Ultra Disks. They were actually able to get more performance, for less cost compared to using Premium SSDs.
To conclude this section, I want to note two design constraints when selecting disks for your VM.
- The VM SKU is still limited to a certain number of IOPS, Throughput and Disk Count. Adding together the total performance of your disks, cannot exceed the maximum performance of the VM. If the VM SKU supports 10,000 IOPS and you add 3x 60,000 IOPS Ultra Disks, you will be charged for all three of those Ultra Disks at their provisioned performance tiers but will only be able to get 10,000 IOPS out of the VM.
- All of the hardware performance may still be subject to the performance of the access protocol or configuration, more on this in the next section.
Additional Reading on Storage:
- Disk Types: Select a disk type for Azure IaaS VMs – managed disks – Azure Virtual Machines | Microsoft Docs
Software Configuration and Access Protocols:
As we come to the last section of this post, we get to the area that aligns with the purpose of this blog series – shared storage. In this section I’m going to cover some of the most common configurations and access types for shared storage in IaaS. This is by no means an exhaustive list, rather what I find most common.
Scale-Out File Server (SoFS):
First up is Sale-Out File Server, this is a software configuration inside Windows Server that is typically used with SMB shares. SoFS was introduced in Windows 2012, uses Windows Failover Clustering, and is considered a “converged” storage deployment. It’s also worth noting that this can run on S2S (Storage Space Direct), which is the method I recommend using with modern Windows Server Operating Systems. Scale-Out File Server is designed to provide scale-out file shares that are continuously available for file-based server application storage. It provides the ability to share the same folder from multiple nodes of the same cluster. It can be deployed in two configuration options, for Application Data or General Purpose. See the additional reading below for the documentation on setup guidance.
Additional reading:
- Storage Spaces Direct: Storage Spaces Direct overview | Microsoft Docs
- Scale-Out File Server: Scale-Out File Server for application data overview | Microsoft Docs
- Setup guide for 2-node SSD for RDS UPD: Deploy a two-node Storage Spaces Direct SOFS for UPD storage in Azure | Microsoft Docs
SMB v3:
Now into the access protocols – SMB has been the go-to file services protocol on Windows for quite some time now. In modern Operating Systems, SMB v3.* is an absolutely phenomenal protocol. It allows for incredible performance using things like SMB Direct (RDMA), Increasing MTU, and SMB Multichannel which can use multiple NICs simultaneously for the same file transfer to increase throughput. It also has a list of security mechanisms such as Pre-Auth Integrity, AES Encryption, Request Signing, etc. There is more information on the SMB v3 protocol below, if you’re interested, or still think of SMB in the way we did 20 years ago – check it out. The Microsoft SQL Server team even supports SQL hosting databases on remote SMB v3 shares.
Additional reading:
NFS:
NFS has been a similar staple as a file server protocol for a long while also, and whether you’re running Windows or Linux can be used in your Azure IaaS VM for shared storage. For organizations that prefer an IaaS route compared to PaaS, I’ve seen many use this as a cornerstone configuration for their Azure Deployments. Additionally, a number of HPC (High Performance Compute) workloads, such as Azure CycleCloud (HPC orchestration) or the popular Genomics Workflow Management System, Cromwell on Azure prefer the use of NFS.
Additional Reading:
- Create NFS Ubuntu Linux Server volume – Azure Kubernetes Service | Microsoft Docs
- azure-quickstart-templates/nfs-ha-cluster-ubuntu at master · Azure/azure-quickstart-templates (github.com)
iSCSI:
While I would not recommend the use of custom block storage on top of a VM in Azure if you have a choice, some applications do still have this requirement in which case iSCSI is also an option for shared storage in Azure.
Additional Reading:
That’s it! We’ve reached the end of Part 2. Okay, here we go with the Pros and Cons for using an IaaS Virtual Machine for your shared storage configuration on Azure.
Pros and Cons:
Pros:
- More control, greater flexibility of protocols and configuration.
- Depending on the use case, potentially greater performance at a lower cost (becoming more and more unlikely).
- Ability to migrate workloads as-is and use existing storage configurations.
- Ability to use older, or more “traditional” protocols and configurations.
- Allows for the use of Shared Disks.
Cons:
- Significantly more management overhead as compared to PaaS.
- More complex configurations, and cost calculations compared to PaaS.
- Higher potential for operational failure with the higher number of components.
- Broader attack surface, and more security responsibilities.
Alright, that’s it for Part 2 of this blog series – Shared Storage on IaaS Virtual Machines. Please reach out to me in the comments, on LinkedIn, or Twitter with any questions about this post, the series, or anything else!
- Part 1: Azure Shared Disks
- Part 2: IaaS Storage Server
- Part 3: Azure Storage Services
- Part 4: Azure NetApp Files (coming soon)
- Part 5: Conclusion (coming soon)