kohera-logo-regular.svg

Hesitating about the cloud will cost you flexibility (and money)

Cloud computing is fully mature, so we believe that many of your SQL Server systems could be running more efficiently in the cloud. We are convinced that a cloud model, whether hybrid or full cloud, is in fact the best solution for most of our customers. The only questions that remain are: when are you going to migrate and with which model?

>>check out our webinar about SQL on Azure

Current on-premises database servers are sized according to their peak capacity, which means that during normal operations they use 3% or less of their CPU power. What’s more, they’re often oversized for redundancy reasons so they can act as a failover system in other instances too. These redundant processors are idling most of the time. That’s a big financial drain if you’re using software, such as SQL, which includes paying a hefty price per processor.

Cheaper operations

One of the key benefits that the Microsoft Azure cloud platform delivers, is the ability to rapidly scale your application to changes in demand. This can be done manually or by using an auto-scaling tool.

It means that if you move to the cloud, you only have to pay for the CPU cycles or processors that you’re actually using. So you could run on a dual core processor for normal operations and temporarily switch to eight processor cores during peak times. In this way, you pay for the six extra ones only for the time you use them.

SQL Server tuning will further bring down your costs. With traditional on-premises computing, few people think it worthwhile tuning the performance of their SQL Server systems. After all, adding extra gigabytes of RAM or additional CPU power is cheaper than investing time and money in a sound database design. But in the cloud, where you pay for what you use, that’s not true. Basic tuning will quickly reduce your operational costs by 20%, while extensive tuning can cut your costs by as much as a half.

Efficient backup

Cloud backup and data recovery is an option you really should consider. Data is at the heart of all organizations and backing up this data is a key part of any business strategy. Cloud-based backup tools are reliable, inexpensive and scalable, while requiring no investment and minimal operational expenses.

Azure Backup is a simple and dependable data protection tool that enables customers to back up their on-premises data to Microsoft Azure. If you’re worried about unauthorized data access, you have several encryption techniques at your disposal.

Azure Backup works efficiently over the network and on your disk. Once the initial seeding is complete, only incremental changes are sent at a defined frequency. Built-in features such as compression, encryption, longer retention and bandwidth throttling help boost its efficiency.

Testing and disaster recovery are significantly cheaper

Currently, most customers still prefer to run their primary servers on-premises and use the cloud for disaster recovery, backup, testing and development. Setting up a really big SQL Server environment to do specialized stress tests for three weeks is something you just wouldn’t do without the cloud. Now you can deploy a low-budget test server, do all the testing and then decommission it when the testing is finished.

The cloud is also great as a disaster recovery solution. You can set it up using minimal funds and resources, let it sync and if your primary site runs into trouble, you can simply upgrade your cloud servers and fail over (which incidentally takes less than ten minutes).

The least expensive way of doing this is by setting up a Windows Server on Microsoft Azure and letting it run your SQL Server disaster licence. This licence is included in your current license if you opted for a Software Assurance licence. But it means that your cloud server will have to act as a passive node. If you want to use it actively, you should make use of a fully licensed system.

Low cost mobile applications

It’s also possible to request a SQL Server database in a Software-as-a-Service model. In this case, you have a database in the cloud with a maximum capacity of up to 500 GB. The costs for these databases are relatively low because you only pay for the resources you use. We think this is ideal for cloud-based applications or applications that need to sync with mobile devices, such as reporting and mobile dashboarding tools. Sometimes you will need to make minor adjustments to make the database compatible, but in most cases all you need to do is package and publish your current SQL Server database to the cloud.

How to fix latency

If your servers are running in the cloud, a good internet connection becomes very important. Otherwise latency can become an issue. If your internet connection is saturated or your latency gets too high, you can suffer from performance degradation.

You should check latency especially if you are running SQL Server in a synchronous AlwaysOn configuration. The straightforward solution would be to run your systems in asynchronous mode, but this has two major disadvantages. There can be no automatic failover between asynchronous nodes because the primary server does not know the state of the secondary one. Even more importantly, in the case of a disaster there is the possibility of data loss (albeit minimal) because the system cannot guarantee that all the transactions are committed on both nodes. These delays can be monitored and in most cases they are within acceptable margins (they rarely exceed five minutes). Still, it’s something that should be taken into consideration.

If these margins are well within your SLAs and expected recovery point objective, the cloud becomes a very attractive solution. But if the margins are unacceptable, you should invest in upgrading your internet connection.
A hybrid model where you depend on your on-premises servers for high availability and run your disaster node asynchronously in the cloud will also mitigate this issue. What’s more, making the cloud node an active partner of your AlwaysOn configuration will give your mobile workers permanent access to their data, even when they cannot connect to your data centre.

The right choice

Although we firmly believe that you should always take the cloud into consideration when designing or discussing your SQL Server strategy, the final decision is still yours. And there is a whole range of options out there. You can go for a full cloud or a hybrid cloud set-up. You can decide whether or not to keep your primary systems on-premises.

Whether you go for virtual servers, databases or just storage space, including high availability and clustering, we are sure that there’s always a solution that’s just right for you. Also, don’t worry about past investments in your infrastructure, as these will still be very valuable.

Finally, there is extra added value if you decide to run your application servers in the cloud as well, because in Microsoft’s cloud model you don’t have to pay for uploads or data transfers between applications that reside in the same cloud.

Parameter sniffing solved with new Parameter Sensitive Plan Optimization feature

If you’re someone that works a lot with Microsoft SQL Server, there’s no doubt that you’ve had issues with an issue called “Parameter sniffing” before. Up until now, you had...

Creating maps with R and Power BI

The possibilities are infinite when it comes to creating custom visuals in Power BI. As long as you have creativity and knowledge about the right programming language, you can let...

Sending monitoring alerts through Telegram

What if you could get the ease of phone notifications for whatever monitoring alerts you need? Then we have a solution for you with the app Telegram. Some of you...

Send mails with Azure Elastic Database Jobs

The DatabaseMail feature in SQL Server and Managed Instance is widely used by many professionals. But what if you want a similar functionality in Azure SQL Database? There are options,...

Sorting matrices in Power BI

Recently I worked on a Power BI project for a client. They had a SharePoint site where they regularly published articles and wanted to pour view data into a report...

The world of data is evolving

The data landscape has changed dramatically over recent years. In the past, we mainly heard that we needed to do as much as possible “cloud-only”—but this trend has become more...