From Cloud computing to fog computing
October 24th, 2017
By Reza Malekzadeh
Moving all these exabytes of data from the Internet of Things to the cloud for analysis purposes won’t be feasible for much longer as this calls for vast amounts of bandwidth which is far too costly. Also, latency is increased as by the time the data makes its way from the network edge to the data center for analysis, the opportunity to act on it might have passed. This is particularly true for time-sensitive decisions such as security issues or when dealing with manufacturing line shutdown for example.
So there is a real need for new infrastructure capable of handling the enormous volumes of data produced each day. Current cloud computing models were not designed for the volume and variety of data generated by the IoT. The first wave of storage was hardware-based and very costly. With companies such as EMC selling storage arrays for over a million dollars. Then we experienced a second wave with the development of software defined storage technologies leveraging more and more commodity hardware and being better adapted to cloud architectures and centralized applications.
We are now entering a third phase of intelligent storage capable of handling larger volumes and varieties of data. Instead of sending all the data to a data center site far from the end-point, we need to be capable of storing and analyzing most data close to the point-of-collection device. This new computing model, Fog computing, allows compute at the edge of the cloud – one could think of it as a collection of multiple proximity clouds. Because it doesn’t need to send all the data to a centralized data center, fog computing reduces data management time and improves quality of service. Additionally, most data is now compressed and encrypted at the source, making it impossible to optimize data transfer and footprint efficiency. By analyzing and assigning the correct value of data at the edge, it is now possible to improve overall infrastructure efficiency and send to the cloud only valuable pieces of information, enabling to take better decisions faster.
We believe that Fog computing calls for the next generation of intelligent object storage, capable of providing all the benefits of traditional storage while running applications at the edge, directly on the storage infrastructure.
We were very impressed by their technology, their team and their achievements. OpenIO offers the most flexible and cost-effective object storage solution to solve the scale-out challenges faced by businesses with thousands of Petabytes. Their open-source solution delivers a new way to build application back-end services, enabling companies to combine storage with integrated data processing on a single platform.
The quality of the founding team played a key role in our investment decision. OpenIO was founded by a small team of experts, engineers, and former intrapreneurs at Worldline, a €1.5 billion Atos company - they are engineers with real field experience and a true understanding of their clients’ challenges and expectations. Their technology does not come out of a research lab but instead came from working with customers and delivering to them a product that met their needs.
We are also very impressed by their recent contract with Dailymotion, the European leader in user-generated and premium video content. Dailymotion chose a small startup which had previously never raised any funds to replace the incumbent EMC. This is proof of the OpenIO excellent execution capability.
We are very happy to welcome OpenIO to the Partech family. We believe that OpenIO will play a key role in solving the data storage challenges of the future. Their total addressable market is projected to reach $4.95 billion. We will be at their side to help them take object storage to a new level.
The need to solve storage problems
Today, more and more devices connect to the internet: smarter sensors are used in many industries to improve output and increase safety, cities all over the world are installing surveillance cameras on every corner and in every car park… New distributed applications and the Internet of Things (IoT) are generating an unprecedented volume and variety of data.Moving all these exabytes of data from the Internet of Things to the cloud for analysis purposes won’t be feasible for much longer as this calls for vast amounts of bandwidth which is far too costly. Also, latency is increased as by the time the data makes its way from the network edge to the data center for analysis, the opportunity to act on it might have passed. This is particularly true for time-sensitive decisions such as security issues or when dealing with manufacturing line shutdown for example.
So there is a real need for new infrastructure capable of handling the enormous volumes of data produced each day. Current cloud computing models were not designed for the volume and variety of data generated by the IoT. The first wave of storage was hardware-based and very costly. With companies such as EMC selling storage arrays for over a million dollars. Then we experienced a second wave with the development of software defined storage technologies leveraging more and more commodity hardware and being better adapted to cloud architectures and centralized applications.
We are now entering a third phase of intelligent storage capable of handling larger volumes and varieties of data. Instead of sending all the data to a data center site far from the end-point, we need to be capable of storing and analyzing most data close to the point-of-collection device. This new computing model, Fog computing, allows compute at the edge of the cloud – one could think of it as a collection of multiple proximity clouds. Because it doesn’t need to send all the data to a centralized data center, fog computing reduces data management time and improves quality of service. Additionally, most data is now compressed and encrypted at the source, making it impossible to optimize data transfer and footprint efficiency. By analyzing and assigning the correct value of data at the edge, it is now possible to improve overall infrastructure efficiency and send to the cloud only valuable pieces of information, enabling to take better decisions faster.
We believe that Fog computing calls for the next generation of intelligent object storage, capable of providing all the benefits of traditional storage while running applications at the edge, directly on the storage infrastructure.
OpenIO will take object storage to a new level
Fog computing is still a relatively new market but it is expected to grow exponentially in line with connected objects. We are very excited about our latest investment in OpenIO as we believe that they are one of the best-positioned players present on the pioneer fog computing market. OpenIO is a Lille-based startup specialized in object storage and server-less computing.We were very impressed by their technology, their team and their achievements. OpenIO offers the most flexible and cost-effective object storage solution to solve the scale-out challenges faced by businesses with thousands of Petabytes. Their open-source solution delivers a new way to build application back-end services, enabling companies to combine storage with integrated data processing on a single platform.
The quality of the founding team played a key role in our investment decision. OpenIO was founded by a small team of experts, engineers, and former intrapreneurs at Worldline, a €1.5 billion Atos company - they are engineers with real field experience and a true understanding of their clients’ challenges and expectations. Their technology does not come out of a research lab but instead came from working with customers and delivering to them a product that met their needs.
We are also very impressed by their recent contract with Dailymotion, the European leader in user-generated and premium video content. Dailymotion chose a small startup which had previously never raised any funds to replace the incumbent EMC. This is proof of the OpenIO excellent execution capability.
We are very happy to welcome OpenIO to the Partech family. We believe that OpenIO will play a key role in solving the data storage challenges of the future. Their total addressable market is projected to reach $4.95 billion. We will be at their side to help them take object storage to a new level.
SUBSCRIBE TO OUR NEWSLETTER
