Predictive Analytics and Machine Learning for Enterprise Cloud
FIGURE 1. Tintri Analytics Dashboard
Ensure your applications always have access to sufficient compute and storage resources, by modeling scenarios and making precise predictions. Only Tintri Analytics—built on Tintri’s CONNECT architecture—ensures you always have the information you need to make the right decisions based on the historical compute and storage usage patterns of all your applications. That’s because of Tintri’s ability to gather data about individual virtual machines and containers at a granular level from both compute and storage. That provides an all-round picture of an application in an instant.
FIGURE 2. Compute and Storage Usage Trends and Projections
Tintri Analytics is a cloud-based SaaS offering that provides organizations the ability to model both capacity and performance requirements. Tintri Analytics uses data lake of up to three years or more of real-time VM- and container-level data, including historical and current performance and resource usage statistics on every application in your environment. All this is collected natively by Tintri CONNECT.
Powered by Apache Spark and Elasticsearch and machine learning, Tintri Analytics then uses these data points to enable powerful predictive analytics and planning capabilities. It uses your application’s historical performance, capacity and working set data to precisely predict your organization’s future needs for capacity and performance for storage and compute. You can even model “what-if” scenarios to assess the impact of changes before they are implemented.
FIGURE 3. Do what-if analysis with Planner
Tintri Analytics leverages machine learning to find hidden insights, without needing explicit programming. In any machine learning model, the quality, quantity and depth of available data, combined with the quality of the model, directly impact the accuracy and usefulness of the predictions. Tintri collects data at the right level of abstraction – the virtualized or cloud application – with up to three years of historical data. The very high quality of these inputs, combined with the power of Tintri’s machine learning algorithms, provide powerful insights and future trend analysis.
Compute and Storage Planning and Forecasting
Tintri Analytics ensures organizations stay on top of their changing needs. It can:
Predicts your needs for storage capacity and performance, as well as compute CPU and
memory as far as 18 months into the future.
Profile application types and then model how changes in application number or behavior
impact your compute and storage footprint.
Model your capacity, performance and throughput to account for organic growth and new
projects in your enterprise cloud.
Conduct what-if analysis to understand the consequences of changes to your enterprise
cloud for next 18 months in advance.
Planning next quarter’s and next year’s needs won’t take a crystal ball—just a few minutes
with Tintri Analytics.
With Tintri Analytics you can develop application profiles based on compute and storage with
just a few clicks. That will help you understand the storage and compute requirements of all
your virtual desktops, SQL servers, Oracle databases or other profiles. You can drill into a
profile to see which specific applications are outliers: well below or above average. Now, when
you get your next request to add applications, you don’t have to guess or deploy and hope;
instead, model the exact impact and make a decision about whether or not you can accommodate
the request with complete confidence.
Tintri Analytics significantly reduces the effort and time spent for resource management and
planning to a few simple clicks.
5 THINGS TO KNOW
ABOUT TINTRI ANALYTICS
Visualize your organic storage
growth based on per-application data
with no guesswork required
Model storage capacity, performance,
CPU and memory requirements
across specific applications
Support multiple hypervisors
Assign each department its view to
manage and plan their applications
and hardware systems with what-if
Simulate what-if scenarios to plan
future needs of your business
You’ll be able to quickly and confidently answer the following questions for compute
and storage planning and analysis:
How is my current resource utilization trending?
Look at the dashboard to see the current consumption of storage
capacity, storage performance, CPU and memory to get an overall
picture of your compute and storage resources.
How is my environment growth trending?
Explore trends and project when you will run out of resources —
understand which resources are your bottlenecks.
What resources are my applications consuming?
Are there any outliers?
Create application profiles and see resource utilization across different
applications. Drill into a specific profile to identify exactly which
applications are outliers and may need special attention.
We are hiring 500 additional employees over the next two quarters.
What additional resources are required?
Use the what-if planner to simulate the load of 500 additional VDI
seats. See the impact on compute and storage and know in an
instant what you’ll need to support this new project.