Servers & Storage

Servers

The type of server you need is dictated by how you plan to use it. It’s important that you think about your current IT structure, as well as how it is likely to progress in the future. Answering the following questions will help to clarify your requirements:

How many users will hit the server?

Simply, the more computers your server needs to support, the higher the performance burden on the server. For example, an entry level server may be fine for up to ten users who perform routine office work, but a company with 100 employees will need significantly more power. The performance of a server is largely dedicated by a combination of the number, and the type of processors, as well as the amount of memory installed. As a rule, if you are planning to expand, you should cater for any future expansion of your business by ensuring the server you choose can be upgraded to support additional users and applications.

Which software applications will be used?

Think about the programs that users will need, and how they intend to use the server. It goes without saying that sending a few emails and occasionally printing a document is a rather different task than, for example, video streaming and database management. You will find that most software packages specify the minimum and recommended hardware specifications they require, including processor speeds, recommended RAM and storage space.

How much data will you have?

It is important to ensure your server has the capacity to store all your existing data, along with that you will generate in the future. If you are looking to estimate the storage capacity you are likely to need, firstly, look at your historical and usage patterns. Answer the following questions: How many files do you store at the moment? How long did it take to create them? How much disk space do they occupy?

Then, use these figures to predict the amount of data that will be generated over the next two to three years. After this, think about how your usage patterns are likely to change with the introduction of your server. Ask yourself whether you will be using new applications, or changing work processes as a result? Finally, don’t forget to add in the space required by the software and operating system you plan to install on the server.

Is your server ‘mission critical’?

Is it vital that your server remains operational at all times? If you know that your business would suffer considerably in the event of a failure, it’s best to consider a redundant solution. This means that secondary hardware components - such as processors, hard disks, fans or power supplies - will kick start if the primary hardware fails. It is also important that you consider the warranty and support service included with your server purchase too. Depending on the level of expertise you have in-house, you may want to upgrade the standard next business day warranty to a timed response instead. 

How much space do you have?

Servers come in many shapes and sizes. We have outlined the main server types below.

1. Tower servers - Built in an upright, standalone cabinet. Similar in size and shape to tower style PCs.
2. Rack servers - Lighter and less bulky than a tower server. Easier to manage in quantity due to taking up less room. Easier to cable. Designed to be bolted on to a rack cabinet.
3. Blade servers - Compact, stripped-down modular servers optimised to minimise space and energy. A blade server fits into a space saving chassis (also called an enclosure or rack) housing similar blades. Primarily designed for enterprise use. 

What about budget?

Once you’ve figured out how you’re going to use your server, you need to ensure you have the right budget for the configuration you need. Be wary of low priced entry level servers which provide limited functionality as standard. You may find that you’ll soon need to upgrade to a more capable model thereby wiping out any initial savings. Whichever server you decide to purchase, the chances are that you will need to upgrade it in line with business growth at some point in the future. Armed with the knowledge relating to your usage requirements, look to spend as much as you can on the hardware and software you need to ensure the server solution you purchase is expandable and therefore fit for purpose both in the short and longer term.

Storage

There are multiple types of storage devices available to suit a wide variety of usage requirements. TPicking the right type of storage solution for your organisation will depend on a number of factors including the volume and type of data you need to store, the type of access you need and of course your available budget. Let’s take a closer look at some of the more common types of storage device available.

Direct Attached Storage (DAS)

A fairly basic level of dedicated storage, DAS consists of a storage device directly attached to your server or desktop PC. In most cases, this device will either be a server itself or an external hard drive. This option is often the cheapest, but it can be limiting for some companies. For example, if data cannot be shared, there are no failover measures. In recent times, however, DAS has experienced somewhat of a revival thanks to increasing virtualisation.

Network Attached Storage (NAS)

This option is more efficient than DAS because multiple network users can share programs, files and other content. With NAS, the server or other hard-disk device holding the data connects to the local area network through a wireless router, not direct to the servers

or other desktops. This means there is always one copy available, in one accessible place. The NAS is configured through a web browser, without any dedicated keyboard or display.

NAS has a number of advantages other than data sharing, however, making it a viable option for businesses. Access is fast because the content is not competing for processor resources. What’s more, power is utilised more efficiently than with DAS. There are also a wide variety of enterprise level solutions available with a good choice of storage capacity, speed and price.

Storage Area Network (SAN)

Not to be confused with NAS (network-attached storage), SAN differs by operating at block level, rather than individual file level. A SAN is a high-speed network of storage devices connected to servers, meaning that applications and data are accessible from any server

on the network. This is in contrast to NAS, which is not an actual network itself, nor simply a storage device. On the whole, SAN is a more sophisticated option than those previously discussed, with high speed fibre channel connections. Compared to DAS and NAS, SAN offers businesses much greater flexibility, availability and performance. Storage is taken away from the servers and consolidated in the optimum place for accessibility. The system also utilises your hardware resources more effectively, saves you buying extra hardware, simplifies management tasks and frees up IT staff for more strategic tasks. There are a wide variety of SAN products available from major IT vendors, including Dell EMC and HPE.

Backup storage

Backup storage solutions help to maintain data continuity in the event of a system failure, natural disaster, or any other form of disruption to operations. There are three main types:

Disk-based backup solutions

This option provides a cost-effective backup storage solution for individual workstations and small servers. It is most suited to individual workstations or small office environments, but requires regular transportation to an off-site location for security purposes. High-speed data protection is just one of the perks of a disk-to-disk backup solution, while reliability is also a key selling point. Using this option, hard drives can be combined using RAID technology so that they work together to store data. Should the drive fail, other drives in the RAID ensure that data is still available. Hardware RAID solutions are commonly implemented through cheaper software, and hybrid solutions are also available.

Other common disk-to-disk solutions include snaps and clones, which - like RAID solutions - protect against data loss and disk failures. However, these have the added benefit of sitting in a separate hardware environment. Disk-to-disk-to-tape backups provide a comprehensive solution by combining automated disk-based backup with tape storage for archiving.

Online and virtual backup solutions

Online and virtual backup solutions are becoming increasingly popular within the SME market due to the fact they provide a simple, inexpensive method for offsite backups. Using this option, a backup agent (which is a small software program) sits on your server and monitors file changes. Then, updated files are transmitted to the chosen service provider’s system via the internet.

While this solution has the benefit of being automated and media-free, it is important to check important services like security, bandwidth and data restoration time frames are suitable for your business. Furthermore, it is worth considering that a degree of customisation control can help you to tailor the solution to match those requirements that are specific to your business.

A virtual backup involves backing up a physical server - such as HPE, Dell EMC or Fujitsu - to a virtual server or storage system. This alternative can help companies cut costs by reducing, or eliminating, the need for IT hardware, power or cooling expenses. A further benefit offered by this solution is that it offers the ability to monitor and manage backups via a web-based console.

Virtual backups are not suitable for every server. For example, an Exchange server under heavy use may require a physical solution. Unless some form of secondary storage is planned offsite, virtual backups are certainly subject to the same natural disaster threats as the original data.

Tape-based backup solutions

A popular backup data solution that’s been around for ages, tape-based storage products are notably fast, reliable and comparatively cheap. What’s more, they are highly portable and offer a high capacity, durable solution that is ideal for archiving applications.

A tape backup device can be fitted directly into a server or workstation, or form part of a dedicated backup system capable of serving multiple servers and workstations on a network. Many small to medium-sized companies use tape either as a primary backup method, or as the final step in a disk-to-disk-to-tape solution.

There are a number of things to consider when choosing a tape-based storage solution:

• Ensure your tapes have sufficient capacity for the amount of data that needs to be stored
• Make sure you know the raw, native capacity of the tapes, rather than the compressed capacity that many manufacturers prefer to quote
• Your data backups need to compete within the timeframe available (i.e. during out of office hours) - so data transfer rate is important

Tapes boast excellent capacities and transfer rates, with LTO tapes available with capacities of 800Gb, and data transfer rates of up to 160Mb/s. There are several solutions available that will help to alleviate the burden of having to manually change tapes. For example, autoloader devices, which can typically store between eight and 16 tapes, automatically load, record and store backups, removing the potential for missed backups and therefore, reducing the risk of data loss caused by damaged tapes. Many tape library systems include management features such as integrated barcode readers.

Tapes are scanned when they are loaded into the library and scanned again when they are removed for storage, simplifying tracking and inventory management. Another time saving feature is web-based remote management software, allowing the tape library to be monitored, configured and managed off-site.