Storing data in the cloud is something that many companies have had to grapple with over the last couple of years. Ranging from the employee who uses it to store files they may want to work on at home or on the move, to enterprise deployments where document management is an essential part of the business.
Whether you are the CIO, IT manager or user, these tips should help make storage in the cloud safe, efficient and reliable.
1 – Know where your data is stored
It seems almost counter-intuitive given that the cloud should mean the data is anywhere and you don’t know where it is, but practically speaking you should be aware of this.
When storing data in the cloud, you need to know where this data is being physically kept to avoid problems such as illegal access and cyber-snooping. This requires you to carry out due diligence as many cloud storage providers don’t disclose which country or legal jurisdiction their customer data is stored.
Not all of these jurisdictions have the regulatory framework in place for data protection and not knowing the true location of critical data could mean you inadvertently breaking the law. Such negligence could land you with a large fine and dent customer confidence.
Many regulated markets require data to be kept within certain countries or regions. If this is the case then you should be able to specify, with their cloud provider, the datacentre locations where data will reside.
2 – What are you going to store?
CIOs need to understand what sort of data they are going to store in the cloud and just how much. This will help you in figuring out the cloud storage provider you need. Sometimes it is better to take small steps by less critical application data in the cloud. Then, as you feel more confident with the cloud storage service, more critical data and apps can be transferred into the cloud.
Be equally aware that not all data needs to or should be in the cloud. Assess what data needs to be where and plan accordingly.
3 – Dedupe and compress backups Despite networking technologies such as fibre coming down in price, it can still be expensive and time-consuming to backup data and store it in the cloud. De-duplication and data compression can ensure that bandwidth is used efficiently and is cost effective. A good backup system can cut down the amount of data needing transfer to the cloud by compressing and de-duplicating it first. Such systems may save thousands in networking and storage costs.
However, it is wise to remember that are also trade-offs as well from backup performance and regulations to bandwidth consideration.
4 – Ask who’s looking after your stored data in the cloud Don't be afraid to ask cloud storage providers what (if any) security is carried out within their clouds - this would enable the identification of any potential violations. Cloud storage providers should be able to describe what types of monitoring policies they have, alongside forensic capabilities, in place to protect customers from any breach and enable any investigation.
You would be well advised to look for providers that conform to standards of good practice ISO 27001 for information security management, ISO 9001 for quality management and ISO 14001 for environmental management. This should guarantee the providers are up to snuff.
Another thing to pay attention to is privileged accounts. If these are compromised a hacker can take control of the cloud and the data stored on it. Make sure security policies and processes apply to data stored here just as they would on-premise.
5 – Secure cloud storage data when in transit
Cloud data has to be secure when in transit; traffic needs encryption using standard security protocols such as AES-256. This will protect the data not only in transmission but at rest as well.
Creating a VPN connection will also ensure that data stays safe as it moves from the organisation to the cloud. A VPN is also useful (if not essential) when connecting an organisation’s datacentre to a virtual private cloud. VPN gateways are also necessary to support remote access to cloud storage from smartphones, tablets and other portable devices.
6 – Seed your cloud data
If you have terabytes of data to be stored in the cloud, it may well be worth seeding this data by physically transporting it to one of your cloud provider’s datacentres rather than rely on moving it through the internet.
This is often the first step in creating a data backup to a cloud service. A cloud storage provider will send a disk drive or appliance to the customer, who backs up data locally before shipping this back to the cloud storage provider. Once this is done, the customer only has to back up blocks of data that have changed following this initial seeding.
7 – Avoid vendor lock-in
A cloud provider may make it easy enough to put data into their cloud and make it accessible, but what happens if you need to move it? Ask the provider how easy it is to move data back out of their cloud. A cloud provider should makes their cloud interoperable with other clouds. Ask the cloud storage provider if they offer APIs to access the service and data, so that the organisations can leave their provider if they want?
For example, cloud services with native OpenStack APIs ensure that users can access the entire OpenStack ecosystem and move data and applications between multiple cloud environments and use third-party management tools.
The CIO should always have a plan to migrate data back to their primary data centre, or to a different cloud vendor, with little advance notice.
8 – Keep an eye on costs, both on-premise and in the cloud CIOs need to be aware of the cost of storing data in the cloud as pricing can vary enormously dependent upon storage type and whether data is tiered.
Some cloud services are great for applications that aren’t frequently accessed but costs can start to mount up significantly when running transactional applications that need frequent access to data stored in the cloud.
Separating data according to operational requirements is essential for keeping costs down. Having low-availability services run on high-performance cloud storage will result in underused hardware and huge bills. Such data separation also has a bearing on the uptime an organisation requires. The difference in price between four-nines availability and five nines is substantial.
Should requirements change be aware that you may be charged extra. Ask why and what for? CIOs need flexibility and may have to adjust storage space, backup policies and bandwidth needs.
Be aware too that some cloud storage providers can charge more to take data out than to put it in.
9 – Optimise the connection to the cloud Moving data into the cloud means the CIO should think about bandwidth and latency. It is a good idea to assess an organisation’s bandwidth requirements and ensure connectivity requirements are met.
The wide area network can have a major impact on cloud storage performance. Low bandwidth, latency caused by long distances and a shoddy network can all affect cloud storage and data throughput. Make sure that technologies such as Wan optimisation is in place, otherwise moving large files to and from the cloud can be frustratingly slow.
10 – Stored data in the cloud can be used for enhancing business intelligence
Disparate data sources within the corporate infrastructure can be moved into the cloud and within the cloud business intelligence tools can be deployed that can fully exploit an organisation’s data. Fully exploiting this data in the cloud will help an organisation by providing it with greater insight into various parts of the business.
