Internet of Things
The thermal management requirements in
edge computing environments
As the growth of the Internet of Things (IoT) gathers pace, edge computing has become critical to information management and the wider infrastructure of the new, interconnected world. Tom Gregory, product manager, 6SigmaET, tells us about the thermal management challenges in edge computing
W
hile many modern devices take advantage of cloud services or data centres, edge computing
signals a step in a different direction. More and more IoT manufacturers and application developers are discovering the benefits of computing taking place at the ‘edge’ of the network, nearer to the devices themselves. The reason for this is that it enables data
to be processed closer to the source of creation, instead of sending it across long routes to cloud servers or data centres. When data processing occurs at the edge of the network, as opposed to in the cloud or at a data centre, it can be analysed and actioned faster. And it’s this growing demand to evaluate and analyse important data in real-time — or as near to real-time as possible — that is fuelling the surge of interest in edge computing. By using edge computing, sensors,
controllers and other such connected devices can either collect and analyse IoT data themselves or transmit it to a nearby device for analysis. This requires significantly lower latency than can be achieved in typical centralised data centres and cloud architectures, where processing power is often many miles away from the user’s location.
By moving computing power to the edge of the network manufacturers and developers can overcome this issue, enabling far more responsive data applications – whether that involves smartphones, sensor networks or industrial machinery.
The shorter geographic distance between computing power and user opens up new potential applications for businesses which rely on rapid response rates. For services that need to respond quickly, low latency is crucial. So, it’s better for data to be stored or processed locally at (or near to) the device level. But, while the technology is
threatening to disrupt colocation and cloud computing, it’s also creating some specific challenges for organisations when it comes to the design of edge servers. The crucial issue is thermal management. For any server, thermal design is important when it comes to ensuring reliability and energy efficiency. But typical server design has been able to rely on devices being housed in large racks, complete with powerful cooling fans, situated in purpose-built data centre facilities with sophisticated climate control systems. Which makes the thermal design challenge of the servers
themselves relatively more straightforward to deal with. In comparison, edge computing servers need to be significantly smaller – restricting natural airflow through the device – and there are also some issues in terms of the external environment. When edge servers are deployed in offices, in factories, under desks, inside street furniture, or alongside pre-existing IT systems, you simply can’t assume there will be an optimal environment for heat dissipation.
Another aspect that needs to be
considered is the nature of the tasks edge servers will be used for – real-time, high performance, low latency – means that downtime will simply not be tolerable. As a result, edge servers need to be bulletproof and any risks from
overheating have to be mitigated at all costs. And with internal space being at a premium in smaller servers, the margin for error is slim.
So, for thermal engineers, it’s important to ensure that designs are right first-time around. And the best way to do this is by using thermal simulation software. When it comes to edge computing, engineers need to prioritise cooling measures and build in tolerances for thermal issues. This might mean redesigning the layout, selecting different materials or choosing different components. Either way, thermal engineers need to be confident that their device will perform under challenging circumstances. By using simulation software, such as 6SigmaET, thermal engineers can test the heat-flow of their design prototypes prior to build. This allows them to visualise how the heat will flow inside the server, how it might affect components and to amend their designs accordingly. As a result, thermal simulation
software, and the access to accurate and actionable test data, is integral to solving some of the thermal challenges posed by edge computing.
www.6sigmaet.info
www.cieonline.co.uk
Components in Electronics
October 2018 31
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68
