After recently reading a post by Alan Quayle on the “Real-Time Cloud,” I thought it might be a good idea to give a general overview of the various aspects of this new offshoot of the Cloud. Put simply, the real-time Cloud is a high-performance cloud infrastructure that supports real-time services and APIs for any device connected to the Web.
agile, fast and adaptable
Before diving into the heart of the matter, I’d like to underline that the Cloud notably aims to improve the agility, speed and adaptability of processes in a distributed, evolving and collaborative world where efficient communication is crucial for innovation.
With this in mind, I’ve outlined four different points of view that are interrelated, complementary and essential to building the real-time Cloud.
the “network” point of view
Network quality, the backbone of any digital communications, is a network’s capacity to deliver maximum bandwidth, minimal lag and optimal SLAs. It is crucial for efficient data exchange between any number of terminals located at any distance from each other. These are our core goals and the main advantage of the Cloud, and I won’t add anything further.
the “infrastructure” point of view
Based on increasingly sophisticated virtualization platforms and hypervisors, the infrastructure portion runs on physical layers such as servers and physical hard disks. With years of performance management experience behind their development, the capacity of these layers can be increased or decreased depending on need. In some cases, they may be powerful enough to process data in real time. Specifications of these layers should be set to meet the target performance for the type of process being set up.
In-memory analytics may also be used to provide faster database transactions. SAP has provided one example of this by introducing HANA and recently upgrading its ERP suite with this technology. With in-memory, database transactions are much faster, since this technology does not need to access physical disks. AWS also released an option to combine EC2 Cloud services with high-memory clusters using in-memory analytics to power high-performance applications.
Lastly, the virtualization portion needs to offer maximum performance, both in terms of system overhead and efficient I/O management. It also has to provide optimal interoperability with the outside world, which may include several different cloud types.
the “business processes” point of view
Whether it’s in a collaborative context shared by several industrial companies working on the same project – as is often the case in the automotive and aeronautics sectors – or in a financial decision-making context based on risk assessment or industrial management, data transfer speeds are essential to ensuring high-quality business processes for individual sites and their interactions.
The APIs needed for data transfer also need to be fast, efficient and able to find the best paths through the network to avoid any delay in processing data.
the “mobility” point of view
The arrival of 4G LTE will expand some professional uses since it will increase bandwidth and significantly reduce lag (in theory, 150 Mb/s and 20 ms of lag). With this new technology (is this the real birth of the mobile Cloud?), connected devices will help expand the workplace by offering complete access to all business services combined with optimal comfort.
Connected devices need to run on a real-time cloud platform equipped with the latest technology such as HTML5, Node.js and NoSQL. One of the innovations made possible by this technology is discussed in an article on the first Telco PaaS API developed by Voxea Labs and Joyent. Telcos can now present Telco APIs to developers in order to build Mobile VoIP services to go along with IMS solutions and new real-time communication technology using WebRTC.
keyword: real time
To wrap things up, the technology needed for innovative uses powered by real-time services is already or will soon be in place for both the Cloud and the mobile Cloud.
In addition to WebRTC, one of these new uses is cloud gaming, which needs maximum bandwidth and minimal lag. We’ll take a closer look at cloud gaming later on...
Gilles
This blog post was originally published in French here.
photo credit: © Africa Studio - Fotolia.com
Passionate about physics and specializing in numerical simulations, I started my career with 3D simulations of turbulent flows and spatial trajectory. To learn more about the parallel computing architecture used in computational science, I worked with hardware manufacturers in the US and Europe, where I saw firsthand the development of grid computing and cloud computing. I am a technical and commercial expert in cloud solutions, currently focusing on Cloud Business Development by working with my network in the IT world.