The COVID-19 global pandemic continues to affect every organization on our planet. We've been seeing the significant impact it has on the private and public sectors, and this weekend brings confirmation that large public cloud providers are also experiencing difficulties.
On March 28, Microsoft shared a new update on its cloud services continuity, with the main points being:
- Microsoft observed 775% (!) increase in Microsoft Teams calling over one month in Italy since they started to observe social distancing
- Windows Virtual Desktops (VDI) usage has increased 3x
- Power BI, Microsoft’s business analytics service, experienced a 42% increase over a week as the government used the service to share information and insights with the public
- Microsoft Teams, its collaboration and communication platform, increased by over 12 million daily users over a week to more than 44 million daily users
A week prior, Microsoft shared that it has established new priority criteria for new cloud capacity in certain regions:
"Top priority will be going to first responders, health and emergency management services, critical government infrastructure organizational use...". Microsoft reaffirmed in its latest update that there are no changes to the prioritization criteria.
Although AWS has not made a similar announcement, AWS also depends on a supply chain for various components to build its hardware.
Microsoft’s public disclosures are proof that the cloud does not provide "unlimited" or provides "infinite capacity" -- it is quite apparent that cloud providers are not immune to supply chain disruptions.
Avoid “Hoarding” Cloud Resources
This conundrum caused by COVID-19 is expected to last from weeks to months, and it is likely to get worse before it will get better – needless to say, this is a concern for every CIO. It is expected that organizations will focus more on reducing OPEX on the cloud and less on cloud migrations if the cloud providers place restrictions on new workloads.
During unprecedented times like today, we always come together as a society to overcome common threats. We must rely on each other to help the vulnerable, support those on the front-lines, and share limited resources among each other.
The disruption of the supply chain is going to last for a while, and the longer it lasts, the cloud capacity will be prioritized for initiatives and projects aimed at battling the crisis.
It is time for us to employ social responsibility when it comes to the consumption of cloud capacity. We must only consume the cloud resources we need when we need them.
Many of us noticed the empty shelves at the supermarkets over the last few weeks. The uncertainty caused some to hoard resources, a lot more than they will ever need, leaving others with bare shelves. There are enough resources for all of us, but we all must buy only the resources we need when we need them. Otherwise, the shortages will impact fellow citizens.
Consuming cloud resources is no different. As cloud resources are becoming scarce, both due to the exponential growth of applications' usage and supply chain disruptions, every organization must take action now to ensure there is enough for others who will need them.
These resources might be needed by research centers to run cloud-based simulations and modeling to find a cure or vaccine. These resources might be required by the government to power the applications and systems designed to help track the virus spread or running the systems that will allow the public to apply for emergency financial support as part of the stimulus package.
Leaving workloads significantly over-provisioned knowingly or leaving non-prod workloads running on the cloud for no reason is, in a sense, a form of resource hoarding.
What can you do today?
Very simple: organizations must embrace cloud elasticity.
Embracing cloud elasticity means:
- Safely scaling workloads down by matching application demand to the cloud compute configuration (e.g., Instance type of SKU)
- Scaling up only when needed
- Suspending non-prod workloads
- Doing all the above, continuously, 24/7
Executing these actions will not only help organizations to reduce their cloud spend during times of financial uncertainty, but also ensure that the public cloud capacity is there when we need it.
It is time to add cloud optimization to the list of social responsibility acts, which include staying at home, exercising physical distancing, and washing hands.
We can help!
Today, more than ever, Turbonomic brings tremendous value and relief to our customers during these trying times:
- Turbonomic is helping customers reduce cloud OPEX while assuring application performance by helping them embrace cloud elasticity, including workload scaling with RI awareness, reservations management, containers optimization, resources deletion and workloads suspension
- Turbonomic is helping organizations to maximize the utilization of their existing infrastructure with workload placement, super-clusters creation, workload rightsizing and ensuring their critical application are performing with continuous Application Resource Management
- Turbonomic is helping customers maintain VDI performance as their workforce has become almost exclusively remote (Click here for details on COVID-19 VDI initiatives)
- Turbonomic is helping essential services organizations (for example healthcare providers, government, emergency services, etc.) who are expediting cloud migrations and assuring application performance
- Turbonomic is providing free downloads to new customers and free remote training to new and existing customers during the COVID-19 crisis
In closing, I would like to quote Ben Nye, Turbonomic CEO, who wrote in his Open letter on COVID-19: “Together, I know we can conquer this challenge and emerge with an even stronger community and purpose.”