Skip to main content

The real difference between an Azure VM with or without SSD

I want to talk about the real difference of an Azure VM with or without SSD. This is not a post with charts and artificial benchmarks; it is just a real story from the field.
Context
One of my fellows from my work came at me complaining about a performance issue related to SQL Server. On an Azure VM with Linux they used to have an SQL Server instance. The DB storage size was not to complex and the DB size was acceptable.
Problem
Every few hours a job has to be executed on the database. There is a lot of data processing inside it and it usually takes around 1 hour. From duration perspective this is not acceptable, there is a clear NFR that request the task to be executed under 30 minutes.
An audit was done to the VM and database and it was pretty clear that there is a problem at read and write operations. Many actions were happening at that level, causing the memory and storage to be at high levels.
Actions
The DB specialists reviewed the database structure and the job. Unfortunately, there was not to many things that they could optimize. In addition, different things were tested like enable/disable different cache level and other SQL Server configurations, but without success.
The IT team also checked the Linux configuration and tried to see if there was an issue with the VM itself, but like in the case of DB nothing relevant was found except that the disk and memory were at high limits.
VM Resources
When we looked at VM type, we discover that it was an A4 with 8 cores and 14Gb memory. In theory this should be more than enough, with plenty of memory and CPU.
When we looked at the storage we noticed that behind the disk there is a normal HDD. This was the first sign combined with the results from DB and  the IT team that reported high disk and memory consumption. Usually, a system like SQL Server tries to use more memory when the disk is too slow and cannot keep up with the load.
Solution
We migrated the VM from A4 to D4S v3 that has only 4 cores and 16Gb memory, but has a powerful SSD behind it offered by premium storage.
Surprise, from more than an hour we were able to reduce the SQL job to 7 MINUTES. WOW! This is a big difference, that was mostly influenced by the storage type.
Less than 50% of memory is now consumed and the boost from SSD makes the job to fly. The funny thing is that we even pay less. An A4 cost us around 255e/month in comparison with a D4S v3 that is less than 150e/month.

Lesson learned
Before deciding what kind of machine you want to use from Azure, try to put on the paper what kind of resources you will need most. Based on this, try to choose the best VM that suites your needs and don’t forget that Microsoft has a good documentation related for different VM types (General Purpose, Compute Optimized, Memory Optimized, Storage Optimize, GPU and High Performance Compute).
And yes, play with different VMs configurations to see what works best for your needs.

Comments

  1. SSD disks for SharePoint Server is a no brainer wither in Azure IaaS or On Prem. At the least the SQL DB server should have SSD.

    ReplyDelete

Post a Comment

Popular posts from this blog

Windows Docker Containers can make WIN32 API calls, use COM and ASP.NET WebForms

After the last post , I received two interesting questions related to Docker and Windows. People were interested if we do Win32 API calls from a Docker container and if there is support for COM. WIN32 Support To test calls to WIN32 API, let’s try to populate SYSTEM_INFO class. [StructLayout(LayoutKind.Sequential)] public struct SYSTEM_INFO { public uint dwOemId; public uint dwPageSize; public uint lpMinimumApplicationAddress; public uint lpMaximumApplicationAddress; public uint dwActiveProcessorMask; public uint dwNumberOfProcessors; public uint dwProcessorType; public uint dwAllocationGranularity; public uint dwProcessorLevel; public uint dwProcessorRevision; } ... [DllImport("kernel32")] static extern void GetSystemInfo(ref SYSTEM_INFO pSI); ... SYSTEM_INFO pSI = new SYSTEM_INFO(

Azure AD and AWS Cognito side-by-side

In the last few weeks, I was involved in multiple opportunities on Microsoft Azure and Amazon, where we had to analyse AWS Cognito, Azure AD and other solutions that are available on the market. I decided to consolidate in one post all features and differences that I identified for both of them that we should need to take into account. Take into account that Azure AD is an identity and access management services well integrated with Microsoft stack. In comparison, AWS Cognito is just a user sign-up, sign-in and access control and nothing more. The focus is not on the main features, is more on small things that can make a difference when you want to decide where we want to store and manage our users.  This information might be useful in the future when we need to decide where we want to keep and manage our users.  Feature Azure AD (B2C, B2C) AWS Cognito Access token lifetime Default 1h – the value is configurable 1h – cannot be modified

What to do when you hit the throughput limits of Azure Storage (Blobs)

In this post we will talk about how we can detect when we hit a throughput limit of Azure Storage and what we can do in that moment. Context If we take a look on Scalability Targets of Azure Storage ( https://azure.microsoft.com/en-us/documentation/articles/storage-scalability-targets/ ) we will observe that the limits are prety high. But, based on our business logic we can end up at this limits. If you create a system that is hitted by a high number of device, you can hit easily the total number of requests rate that can be done on a Storage Account. This limits on Azure is 20.000 IOPS (entities or messages per second) where (and this is very important) the size of the request is 1KB. Normally, if you make a load tests where 20.000 clients will hit different blobs storages from the same Azure Storage Account, this limits can be reached. How we can detect this problem? From client, we can detect that this limits was reached based on the HTTP error code that is returned by HTTP