Ontap AI: A New Direction for A.I.
It takes more than complex Algorithms to create fully functional Artificial Intelligence. The data on which they are built on is just as important; if the data is not reliable, your algorithms are operating on false pretences. According to the IDC (International Data Corporation) there will be a little over 44 Zettabytes (1 trillion Gigabytes) of digital data by 2020.
This poses numerous challenges including: security, Big Data Costs, heat diffusion and energy consumption (It is estimated that 1.1% - 1.5% of the world’s total energy in 2010 was used to run data centres). Thankfully, due to rise in storage technology, management tools and cheaper media costs, data storage costs are declining steadily.
Processing, collating and transferring AI model data is difficult. As such not all cloud storage providers guarantee the same level of service and reliability.
Nvidia and NetApp saw this difficulty as an opportunity to reinvent the way we develop AI.
Boasting Nvidia’s technological know-how, their new flash storage is designed to help individuals and organisations achieve “edge to core to cloud” control over their data.
NetApp’s SaaS solution, Data Fabric, excels in unifying data sources across clouds (clouds in Data Centres, public cloud offering from service providers and hybrid/pubic clouds). It provides immediate access to data regardless of its format or physical location.
Thanks to these giant leaps in cloud computing, organisations can benefit from seamless data transferring and more complex systems such as Artificial Intelligence which requires advanced algorithms and vast, immediate and accurate data processing.
For us individuals, we can look forward to a better service and overall experience with AI.
1.Wahlroos, M., Parssinen, M., Manner, J. and Syri, s. (2017). Utilizing data center waste heat in district heating e Impacts on energy efﬁciency and prospects for low-temperature district heating networks. [ebook] Aalto: Elsevier Ltd. Available at: https://reader.elsevier.com/reader/sd/38173360338CC4873ED2D45036C692B88F502144BC86BE4056A59FC16383C630C46520750146AF97F3F86816910862CC [Accessed 23 Aug. 2018].