How EkkoSense is leveraging the latest Machine Learning and Artificial Intelligence capabilities in its EkkoSoft Critical technology to help organisations unlock carbon savings in their data centres
EkkoSense Ltd
Data Centre Intelligent Automation and Management Innovation of the Year
Entry Description

Brands know they need to be targeting significant energy optimisation results if they are to achieve or even get close to their net zero carbon commitments. Data centres have a critical role to play here. They already represent an organisation’s second largest consumer of energy, with around 35% of this taken up by powering cooling equipment. Therefore, any initiative that can reduce data centre cooling will help to make a significant contribution towards their carbon reduction goals.

Organisations clearly want to deliver on their carbon obligations, but that can be challenging when data centre operators don’t have a clear understanding of how their rooms are actually performing from a cooling, capacity and power perspective. Indeed, when faced with an external issue – such as an increased thermal demand placed on facilities by a surge in hosted services, for example, - the default position for many operations teams remains to keep throwing more cooling at the problem. This simply adds to the data centre’s overall carbon footprint, and often does little to resolve the original issue.

So, while operators remain keen to secure carbon reductions, the reality for many is that they don’t have access to the tools that could help them make smart choices. With the ongoing pandemic impacting investment decisions, many also don’t have access to capex funding to address this. And, while data centre operators can deliver unprecedented efficiency and environmental performance levels at hyperscale, the reality is that most data centres still face considerable operational challenges. That’s where EkkoSense and its EkkoSoft Critical software-driven optimisation solution for critical data centres can help.

What is novel about the EkkoSense EkkoSoft Critical approach?

EkkoSense is the only organisation to directly address the fundamental challenge of allowing data centre operations teams to gather and visualise cooling, power and space data at a granular level.

EkkoSense does this by bringing together an exclusive mix of technology and capabilities – including a unique SaaS platform, low-cost Internet of Things (IoT) sensors, and PhD-level thermal skills – to facilitate the crunching of multiple complex M&E datasets to help data centre teams support instant optimisation decisions.

The result is a 3D visualisation and analytics platform that’s particularly easy for operations staff to use and understand. This lets them visualise airflow management improvements, manage complex capacity decisions. and quickly highlight worrying trends in cooling performance. Having this level of immediate insight means that our customers can:
• Deliver proven data centre cooling energy savings – 30% on average
• Reduced energy usage unlocks immediate carbon savings
• Software and optimisation costs covered by reduced energy spend
• Software-driven process unlocks additional benefits including
o Freeing valuable data centre capacity
o Ensuring 100& rack-level ASHRAE or SLA thermal compliance
o Remote optimisation capability supports socially-distant onsite optimisation activities
Despite data centre operators continuing to make significant investments in IT infrastructure and management, their critical facilities continue to face significant operational challenges. Thermal issues rank as today’s second largest cause of data centre loss of service - accounting for 29% of unplanned outages. And, while 35% of data centre energy consumption is already attributed to cooling, overall cooling utilisation still only averages 38%.

The result is that many data centres remain exposed to thermal risks, with their operations teams often unaware of the scale of the problem they face. Too many still rely on legacy BMS systems to highlight thermal risk, or they hedge against potential temperature problems by routinely over-cooling their facilities. This outdated approach not only fails to address the underlying issue of data centre thermal risk – meaning that almost all data centres aren’t 100% ASHRAE thermally compliant - but also does little to help the wider organisation as it struggles to reduce overall business carbon footprint levels.

How does the EkkoSense approach to unlocking energy savings break with conventional ideas/processes in its field?

EkkoSense’s AI and machine learning led approach directly addresses key data centre challenges such as excess cooling energy consumption, releasing stranded capacity and identifying thermal risk leading to unplanned outages – all at a fraction of the cost of expensive and complex DCIM or CFD solutions. EkkoSoft Critical is able to analyse multiple complex data sets simply and quickly in a 3D environment and apply AI and machine learning analytics to provide actionable improvements. Once actioned, this process can be repeated and measured for continuous improvement – 24x7x365.

Increasingly, this kind of AI & ML powered optimisation platforms will become the norm – and even more pervasive. EkkoSense believes this is just the beginning of what these technologies can offer the data centre environment.

Creating a Digital Twin of your data centre layout

Combining the power of artificial intelligence with real-time data from a ‘fully-sensed’ room enables the creation of a digital twin of an organisation’s data centre layout – one that not only visually represents current cooling, power and thermal conditions, but also provides tangible recommendations for optimisation. This level of decision support can help operations teams to take things to the next level, as the software will recognise when changes have been made and even refresh the digital twin to reflect updates.

So instead of simply automating systems and trusting an AI to get on with managing the sensitive security and controls needed for critical data centre cooling duty performance, EkkoSense believes a more productive approach is to gather cooling, power and space data at a granular level, visualise that complex data to make it easier to compare changes and highlight anomalies, and then use machine learning and AI to provide the actionable insights your data centre team need to do their jobs more effectively. With its 2020 edition of EkkoSoft Critical, EkkoSense has now put this into practice with Cooling Advisor, the industry’s first advisory solution embedded within a data centre optimisation solution.

How does It go beyond marginal improvements on something that already exists?

Built right into the heart of the EkkoSoft® Critical SaaS 3D visualization and analytics software, Cooling Advisor provides data center operations teams with focused cooling performance recommendations and advisory actions.

By offering focused cooling performance recommendations and advisory actions, the new Cooling Advisor capability helps organisations to unlock significant further data centre cooling energy savings. Just by acting on the software’s advice alone. And these actionable changes – such as suggested optimum cooling unit set points, changes to floor grille layouts, checking that cooling units are running to specification, fan speed adjustments or advice on optimum rack locations – are all presented each time for human auditability. Data centre operations team members make the suggested changes, and then use the software tool to confirm that the Cooling Advisor recommendations are delivering the expected results.

Unlocking significant data centre cooling energy savings

On top of the 30% average data centre cooling energy savings delivered through optimisation using EkkoSoft Critical, the Cooling Advisor advisory solution that’s now embedded in the software equips data centres with an additional self-optimisation capability. This enables organisations to benefit from continuous, year-round performance optimisation, with benefits including:
• Guidance on Cooling Set Point Adjustments
• Highlighting of under-performing or redundant cooling units
• Identification of floor grilles that should be removed or relocated
• Allowing operational teams to take control of their data centre optimisation
• Uncovering of critical cooling Zones of Influence using machine learning techniques
• Support for dynamic critical facilities, helping rooms to remain fully-optimised
By deploying Analytics and Machine Learning, EkkoSense has enabled its software-driven approach to give daily advice on how to optimise data rooms. The resulting cooling savings from this can pay for the platform very quickly (<1 year). The advice is specific to every facility, it is not generic advice. On average this has delivered a 10% - 15% reduction in cooling energy cost out-of-the-box.

In adopting this approach we’re also recognising that data centres never stay the same, so this kind of ongoing optimisation allows teams to keep up with their ever-changing environment. Additionally, by deploying cooling duty sensors to track cooling loads, we’re also now able to identify previously undetected cooling unit faults before they are even picked up by BMS alerts.

Creating live airflow ‘Zones of Influence’ for even greater control

By capturing entirely new levels of data centre cooling data – going beyond basic temperature measurements to also include energy usage and airflow distribution – we’re able to map zone-by-zone cooling analytics within the data centre to help with resiliency and capacity planning decisions.

This creates live airflow ‘zones of influence’ that can group racks into clusters that specifically match the cooling unit that’s provisioning them. It also supports previously unattainable measures such as capturing airflow within data centre floor vents. The application of analytics and machine learning here shows which AHU units are serving which racks, and with Cooling Advisor will show precisely which AHU could be put into standby or turned down if using variable speed fans.

With this kind of real-time thermal monitoring in place, data centre teams can track cooling outputs and identify any poorly performing cooling systems in advance so timely improvements can be made. EkkoSense supports both granular rack and CRAC level monitoring in order to find the hidden – but easy to fix – cooling and airflow problems that typical cooling PPMs and BMS systems fail to find or diagnose.

EkkoSense - Delivering tangible improvements and measuring their effect

It’s only by removing 100% of thermal risk from their data centre operations and providing a stable platform for subsequent cooling optimisation projects, that data centre managers can start to build truly optimised data centre rooms. Unlike traditional BMS or DCIM systems, EkkoSense offers all the elements needed to reduce risk and cut data centre cooling energy costs – rather than pursue the traditional reactive approach to hard faults.

EkkoSoft Critical also recommends actionable changes, for human auditability – such as set points, floor grille layouts, cooling unit operation, fan speed adjustments and optimum rack locations – and it can also be responsible for fully autonomous changes. At all times, live, real-time visualisations are critical in order to provide immediate performance feedback on actioned changes.

Given that data centres are constantly evolving, this level of continuous optimisation is critical in accommodating inevitable changes. As a SaaS-based visualisation and analytics tool, EkkoSoft Critical is easy-to-use either in-situ or remotely, is easy to maintain and encourages a standardised, consistent approach across multiple sites.

How do customers benefit from the product/process or innovation?

Interxion – a leading European provider of carrier- and cloud-neutral colocation data centre services (and a Digital Realty company) – has worked with EkkoSense and EkkoSoft Critical this year to reduce its cooling system energy consumption by 20% during the first 12 months of an ongoing project.

The software provides real-time visibility of Interxion’s key data centre operational and thermal performance to monitor one of its key London sites, identify areas for improvement and recommend appropriate changes to optimise performance. EkkoSense’s software-driven insights have been the key driver behind the success of the project, reducing cooling energy consumption for Interxion and supporting the company’s goal of creating green, energy-efficient data centres.

The unique performance analytics built into EkkoSoft Critical have highlighted airflow and cooling inefficiencies, and provided visibility into critical metrics throughout the energy-saving process. The Interxion team now uses the software on a daily basis for ongoing monitoring and evaluation capacity planning activities to maintain an optimised state as its data centre environment changes. The project with EkkoSense is expected to expand to two New York data centres operated by Digital Realty as well as five additional European sites.

EkkoSense solutions have now been deployed in over 350 data centres, with over 100,000 racks and 4,000 cooling systems now globally monitored and analysed 24x7x365.
Supporting Documents