AWS expands edge computing service to Singapore

Amazon Web Services (AWS) has rolled out its Snowball data transfer and edge computing service in Singapore, underscoring efforts by the cloud behemoth to ride on the growing momentum of the internet of things (IoT) and hybrid IT in the Asia-Pacific (APAC) region.

With more IoT devices expected to be deployed across the region, enterprises will need access to computing and storage capabilities at the edge of the network to crunch data and speed up decision-making, without having to rely on the cloud.

AWS’s Snowball Edge, also available in Mumbai, Tokyo and Sydney, lets enterprises do just that. With a Snowball Edge appliance, enterprises can hold up to 100TB of data at the edge of a network, as well as encrypt and process all that data using AWS Lambda serverless functions through the AWS Greengrass service. 

The ruggedised appliance, which will work in tandem with other Snowball Edge devices to form an on-premise storage pool, can be shipped back to AWS, which will load the data into the Amazon Simple Storage Service.

Since its launch in 2016 in the US, followed by other regions including the European Union and South America, Snowball Edge has been used by the offshore marine industry to process data while the device is disconnected to the network, among other use cases.

“They could use it with seismic and geophysical data gathered by remote sensors that are not connected to the AWS cloud during the period of operation,” said Ian Massingham, AWS’s global head of technical and developer evangelism, at the AWS Summit in Singapore. “When they bring it back, the data would have already been cleaned up and transformed.”

However, the Snowball Edge appliance is relatively bulky, making it unsuitable for a “lightweight, connected device context” where it may need to go into a vehicle and be moved around. Massingham said that in such situations, enterprises can choose to deploy the AWS Greengrass software on smaller devices, such as the Raspberry Pi.

Through the use of Greengrass Core – the runtime that enables local execution of AWS Lambda, messaging and security – these devices could serve as the core of an IoT network and communicate with other IoT devices through MQTT, a machine-to-machine connectivity protocol.

Global mining giant Rio Tinto is already monitoring its fleet of autonomous vehicles in real time using Greengrass.

Massingham said each vehicle is equipped with microcontrollers and custom software that connects to other sensors in a mine’s physical infrastructure to detect issues with mining equipment. Once a vehicle passes by a Wi-Fi hotspot, Greengrass Core will offload the data to the AWS cloud for analysis.

When necessary, enterprises also can deploy cloud-trained machine learning (ML) models through the Greengrass ML Inference capability on a local device, so that an application can quickly respond to local events.

Industry analyst firm IDC said Snowball Edge marks the start of a new direction of growth for AWS’s infrastructure offerings, focusing on computing and storage needs at customer premises.

“The general availability of the VMware Cloud on AWS further strengthens AWS’s level of support for hybrid use cases – providing interoperability and integration across public and on-premise platforms across a wide range of use cases,” it said in a September 2017 research note.

Massingham said adoption of VMware Cloud on AWS, currently available in the US and the UK, has been excellent, mostly by VMware customers that can take their expertise to the cloud without the need to operate a datacentre.

VMware Cloud on AWS is expected to be available in APAC by the end of 2018, at prices that will be set based on market differences such as the scale and maturity in adopting virtualised infrastructure, according to Duncan Hewett, VMware’s senior vice-president and general manager for the region.