International. Fujitsu has presented its predictions in the field of Data Center for this year 2019. The Japanese multinational sees 6 fundamental trends that will mark the evolution of Data Centers and that are based on the optimal management of large volumes of information, the impulse of quantum computing, machine learning and sophisticated cooling systems.
1. Non-Volatile Memory Express (NVMe) is the future of flash storage. Enterprises have to manage large and growing volumes of information driven by the adoption of data-intensive technologies, which continue to generate more and more parallel data flows. The new NVMe host controller interface and storage protocol can process up to 64,000 parallel data streams, compared to SATA or SAS protocols that can only handle one queue at a time. Fujitsu has already announced plans to adopt native NVMe combined with PCIe-attached solid-state drives, in its future storage product lines, along with Eternus AF ALL-Flash Arrays, which offer massively parallel, previously unattainable data transfer speeds. In 2019, this huge boost in bus speeds will allow customers to manage massive access to parallel data while ensuring minimal latency. While ultra-high-speed NVMe interfaces exist primarily "within the box," with NVME over Fabric (NVMe-oF), the potential impact of NVMe is no longer limited to accelerated SSDs, but will feed new performance thresholds across servers, fabrics, storage, and system management.
2. We will see the take-off of quantum technology and Fujitsu Digital Annealer. It seemed like an arcane idea just a few years ago, but the reality is that Quantum Computing is progressing rapidly and represents the future of computing, without a doubt. Fujitsu Digital Annealer's novel approach, a quantum-inspired technology-based solution, offers the ability to solve combinatorial optimization problems, which simply wouldn't be possible using conventional computer architectures, at least for the lifetime of a human. This includes examples such as solving problems for: finding similarities in molecules for the discovery of new drugs or materials, optimizing portfolio risk in financial services, safeguarding against a possible financial crisis, increasing the efficiency of robot production, optimizing the warehouse by reducing production time or as optimization of the factory and improving traffic routes in real time for public transport planning.
Far from being a theoretical experiment, Fujitsu Digital Annealer is already in operation to help customers solve challenges such as finding molecular similarities in drug research, reducing production time for custom manufacturing, and optimizing the arrangement of stored components for factories and logistics. In October, details were announced of a proof-of-concept (PoC) project with NatWest in the UK to solve some of the most complex, challenging and time-consuming financial investment problems. It is done through the combination of high-quality liquid assets, including bonds, cash and securities.
3. Machine learning will be accelerated by digital annealing. Based on the quantum-inspired approach, it is machine learning in which all the hard work is done to create an Artificial Intelligence. The process begins by teaching an AI. If you want him to identify a cat, you must show him millions of photos of cats. This process takes time, and if it doesn't work properly, you need to start over. Fujitsu Digital Annealer can help speed up this training process, for example by offering partially pre-trained components such as speech synthesis recognition.
4. A new type of memory will revolutionize server workloads. NVMe technology not only influences the future of flash storage, but also plays an essential role in the development of the next generation of servers. Fujitsu is Intel's cooperation and development partner, responsible for the development of the prototype server system based on Intel Optane DC persistent memory, within the EU's Horizon 2020 project, called NEXTGenIO. This prototype system, which is based on Primergy servers, takes advantage of "non-volatile memory with persistence", which, unlike traditional Dram, combines the speed of main memory with the persistence of conventional storage. By enabling affordable system memory capacities with this new class of memory, end customers can better optimize their workloads by moving and keeping large amounts of data closer to the processor, minimizing the increased latency of retrieving data from system storage. This new technology will be available in the next generation (M5) of the dual-and quad-socket Primergy, as well as in the 8-socket Primequest 3800 series servers, which will launch in April 2019.
5. Container-based use cases will be an accelerator of Hybrid IT adoption. You hear a lot about containers and in 2019 it will increase. It is an aspect of the so-called "micro-service" approach, in which an application is broken down into small, atomic, encapsulated entities with a limited functional scope. These containers can then be dynamically re-composed into complex applications and distributed to multiple platforms, allowing simultaneous processing of multiple tasks. The reward is that micro-services simplify development (which generally enables super-agile DevOps) and maintenance, improve responsiveness to changing demands, enable scalability across individual service levels, and contribute to profitability.
Because containers make it easy to focus on new features, troubleshooting, and shipping new or updated software, they are vehicles for delivering customer value. We implement the use of containers where it meets the demands of use cases, through open source software, such as OpenStack, Docker and Kubernetes, as well as through VMware and Microsoft environments, complemented by Fujitsu's top-layer software products, such as ServerView Cloud Monitoring Manager and Enterprise Service Catalog Manager. And, as a multi-cloud service provider, we also deploy it in our partner clouds, such as AWS and Microsoft Azure.
6. Liquid Inmmersion Cooling systems will produce a breakthrough. Getting the most out of AI or Big Data requires increasing performance of IT systems. Accelerator chips, such as GPUs, require increasing amounts of power, resulting in a dramatic increase in operating temperatures and the need for more cooling. In some cases, data center air conditioning systems are already inadequate, putting limits on how we can exploit the full potential of high-end systems. Fujitsu's Liquid Inmmersion Cooling Systems technology is a solution that uses efficient fluid-based cooling, reduces the demand for large-scale air conditioning equipment, and decreases total energy consumption by around 40%. This reduction in power consumption means that customers in data centers with limited power supplies can roughly double the density of their server and therefore its computing power. Either way leads to big savings in 2019.
