Accessing new resources has never been easier than today with public cloud providers offering compute power on an hourly based with varying specs such as GPU, RAM, CPU, etc. However, recent articles over the web have come to the same conclusion, companies are wasting money in the cloud.
“RightScale reveals that cloud users could reduce their spend by an average of 35 percent by making better use of resources.” BetaNews
“The benefits of shifting business applications to Web-friendly cloud services is proving far more complex than lining up a partner and flipping a switch” Wall Street Journal
“More and more companies are migrating to the cloud for computing power, but many are actually wasting too much money on unused services.” CFO.com
To quickly summarize, the main reasons behind these cloud leaks mentioned across these articles are oversized resource, resource not required at all time and algorithm not resource aware.
7 Recommendations and Best Practices
1. Governance, Time Bounded Resource by Default
By default, AWS, Azure, GCP, Softlayer, etc. give full control to their user to manually select when they want to start and stop resources and services. With a self service portal, IT will enable a selection of services with suggested best practices. For instance, it could include a termination date by default for all AWS EC2 instance created.
2. Automation, Reduce Setup Time
For long term projects, closing instances on a regular basis to save infrastructure cost requires more setup time. With an automation tool including a user-friendly workflow studio, users will quickly build their own setup tasks and schedule them when required. The state of the machine will automatically be restored when needed.
3. Cloud Bursting
To cope with variable, heterogeneous and unpredictable workloads, the infrastructure should be dynamic and adjust its resources in real time to support the load. With this principle, only a minimum amount of resources are required at all time. The application managing the infrastructure will then bursts into a public cloud when the demand for computing capacity spikes and will avoid oversized resources.
4. Resource/Cost Aware Applications
Applications are becoming more and more complex, resource consuming and integrated with others. Moreover, these applications would require different resource types at different time (see Monte Carlo Article). By decomposing these applications into tasks (smaller unit of work), it is possible to associate each of them with relevant resources. Advanced schedulers will then allocate matching resources to each task and distribute the application as appropriate to optimize parallelization of tasks and resource utilization.
“If you build applications that assume the machines are ephemeral and can be replaced in a few minutes or even seconds, then you end up building an application that is cost-aware,” said Mr. Cockcroft
5. Resource Overview
As stated in the previous section, resources are heterogeneous: Ram, CPU, FPGA, GPU, Bandwidth, etc. A level of abstraction is consequently required to offer IT departments a unique dashboard to analyze resource utilization and take appropriate actions.
6. Resource Agnostic
Hybrid and multi cloud infrastructures are gaining interest in many companies since the pros and cons of each system (HPC, private cloud, public cloud, servers, etc.) can be leveraged and companies can avoid lock in. A solution which split the application distribution and resource allocation with the management of resources will enable users to become cloud agnostic. They will be able to answer ecosystem changes with greater agility and leverage the strengths of each resource type and provider.
7. Governance, Good Practice Sharing
Self service portal also has the advantage to reduce IT requests from business users, reduce shadow IT and include an overall management dashboard to track resource utilization and educate about good practices while encouraging innovation.
In conclusion, cloud wastes are due to various reasons and multiple solutions are available to start reducing them.
ActiveEon through its portfolio of solutions and features has tried to tackle cloud wastes applying 7 principles.
“This has the net effect that the Full Reports are ready for viewing 5-8 hours earlier! The overall gain is over 30%.”, ActiveEon's customer Lead Integration Engineer
ProActive Cloud Automation offers a custom portal that IT departments can leverage to suggest and teach best practices. ProActive Workflows and Scheduler includes a workflow studio to easily create automation processes to reduce human intervention, connect dependent tasks and associate specific resource requirements for each task. It also includes a Resource Manager which enables dynamic provisioning of nodes according to various policies and collect metrics for future analysis. The scheduler itself orchestrates the needs of each workflow with available resources to ensure fast, reliable and cost efficient results.
Finally, with ProActive, companies are enabled to target in parallel multiple source of waste leading to greater cost savings.