With servers came Server Sprawl and with VMs came VM Sprawl. So, it only makes sense that with Public Clouds there will also be efforts to maintain growth and cost control.
This typically comes in these three models that all require good Public Cloud inventory documentation:
Converting:
Resources in the Public Cloud can be altered essentially on the fly. Often, administrators will want to convert instances to larger or smaller configurations to help with performance or billing issues.
Migration:
Wanting to move from one Public Cloud to another or from tier to tier inside the same provider.
Repatriation:
It’s common that companies will form a “Cloud-First Strategy” and that might result in the migration of workloads to a Public Cloud provider. Sometimes this results in incompatible results such as unforeseen costs or limitations on performance. In these cases, the administrator might want to repatriate the workload back to an on-premise environment.
Live Optics is deploying a full suite of Cloud-based tools to help you know where you stand with very little effort.
Pro Tips:
For those in the Public Cloud evaluation stage, please check out our Cloud Pricing features that will help you gauge a workload’s cost in the top 3 Public Cloud Providers before you pay the bill!
Also, before moving any workload to a public cloud provider, make sure you read our Right Sizing entry to avoid the most commonly encountered mistake when migrating a VM from your on-premise environment.
Live Optics is including AWS, Azure, and Google Cloud* profiling features as part of our suite. Let’s take a look at the basics of how this works and what to expect for outputs.
Typically, a server inside the environment where resources need to be profiled would host the Live Optics collector, but with hyperscalers all exposing Public APIs and being subscription-based by default, the Live Optics Collector can actually run from anywhere that has internet access.
As a best practice, the administrator would create a read-only account for temporary use.
Profiling a public cloud is actually very fast and this is largely due to the efficiencies of the Public APIs -- you might expect to spend 30 seconds to a few minutes while getting inventory data.
Note: Live Optics will also support the retrieval of performance data and that of course will impose greater amounts of data collection. We are forecasting 5-30 mins for full performance gathering, depending on the size of the deployment. Feature estimated to ship 2nd half of 2021.
Many only use the Public Cloud to host VMs or what is referred to as an Instance. However, there are many offerings that can be obtained that are beyond a Virtual Machine. Live Optics is focused on the most prevalent of these services, but not all.
Breakdown by Provider:
AWS (download the flyer)
- EC2 Instances or Virtual Machines
- S3 or Storage
- S3 Glacier or Cold/Archive Storage
- EFS or Elastic File System
- RDS or Relational Database Services
A full glossary of data points collected and their definitions can be found here. The release notes are also excellent resources.
Azure
- Azure VMs
- Azure Blob
- Azure Archive
- Files
- Microsoft SQL DB Instances
*Google Cloud
At the time of this article’s posting, profiling Google Cloud is not yet generally available. Be on the lookout for updates as this gets released.
There are three models for data collection:
Offline Inventory:
If your needs are limited to just inventory neatly broken down into an Excel document or if you want to preview the results of the discovery, this option is a non-transmission option that leaves the information with you at the conclusion of the collection run.
Live Stream:
Offline is fast and effective but limits your outputs to the basics built into the collector.
Live Stream means that data will be sent to the Live Optics portal in real-time via an encrypted transmission. This will result in processed presentations built in PowerPoint that have easy-to-consume summaries and graphs.
Offline Upload:
This option is very similar to Live Stream... but without the Live Stream! Essentially, the collector will bundle the collection contents into a Live Optics proprietary file type called an .SIOKIT which one can upload later. There is also the option to save your Live Stream collections in this format and if for any reason you want to re-upload your file for report generation, this would be the correct option.