Businesses around the world are set to hugely increase their investments in hardware and networking solutions to support big data analytics operations in the coming years, as there is a growing recognition that more robust solutions will be required in order to cope with the huge influx of data.
Research by Technavio predicts that between 2015 and 2020, global spending on servers, storage and networking equipment to support big data activities is set to increase at a compound annual growth rate (CAGR) of 33 per cent, reaching $26.95 billion (£18.48 billion) by the end of the forecast period.
The biggest sub-sector within this market will be for storage tools, which is set to grow from $3.99 billion in 2015 to $17.6 billion by 2020. The study noted that big data storage will require a huge amount of data handling capacity. In particular, solutions that offer a high input/output per second rate will be essential in ensuring that big data analytics activities can be conducted quickly and effectively.
Technavio noted the major types of storage used in big data are network-attached storage and clustered network-attached storage. Both of these are particularly useful when it comes to storing large quantities of unstructured data.
Amrita Choudhury, a lead analyst at Technavio for automatic identification systems, added: "Introduction of object storage is a new trend, which is expected to drive the global storage market over the next four years. It is an architecture that considers data as objects and block storage, instead of file hierarchy."
The need for improved big data analytics solutions is also a key driver for the server and networking markets, as many enterprises are focusing on increasing their performance in this area in order to keep pace with their competitors.
There is also a growing realisation that data is now a key factor in determining a company's market value, which is driving the adoption of analytics tools and supporting systems in many countries around the world.
Servers that feature direct attached storage are the most common solutions for companies that are thinking specifically about how their infrastructure can better support data processing activities.
"The servers possess features such as high compute intensity, better virtualisation capabilities, modular design, scaling capacity, highly efficient memory, and processor power utilisation, which make it an inevitable part of big data structure," Ms Choudhury stated.
Global spending on server solutions for big data is expected to reach $6.6 billion by 2020, while networking tools, including cabling, routers and switches, with grow by 36.49 per cent CAGR to reach $2.75 billion.