Monitor Server based Storage Assets

What is SMARTDriveAI?

According to Microsoft and Alibaba, servers account for over 80% of data center hardware failures and over 80% of server failures are due to internal drives. SMARTDriveAI is an enterprise class SaaS product that continuously monitors drives within servers. SMARTDriveAI uses advanced analytics and Machine Learning (ML) models at scale to alert IT operations when drive anomalies are discovered. The analytics and ML models were created by observing the health of over 500,000 drives. SMARTDriveAI also provides dashboards for root cause analysis and fault isolation. SMARTDriveAI is like having a drive expert on staff 24x7x365.

SMARTDriveAI lowers the risk of downtime and increases ROI for IT Operations.

Can I try SMARTDriveAI?

Yes, select the 'Start 30 Day Free Trial' button and fill in the form. You will get an email response within a few minutes with instructions on how to start the free trial. If you have any issues or questions, contact the Jeda Analytics team at support@jedaanalytics.com. They will be happy to assist you.

How is SMARTDriveAI different from Datadog, Splunk and other IT instrumentation solutions?

SMARTDriveAI can be thought of as a server based storage focused enhancement to IT monitoring solutions. SMARTDriveAI is a complete SaaS service that's easy to use with minimum user configuration. Other solutions collect data and require custom alerts and custom or template dashboard configuration. SMARTDriveAI contains predefined dashboards and alerts based on the specific purpose of server drive health and performance monitoring. In addition, SMARTDriveAI has embedded advanced analytics and Machine Learning models trained on over 500,000 drives and their behaviors. Other solutions require the user to create these analytics.

Other solutions have more onerous billing, charging on storage as well as devices that are difficult to assess. SMARTDriveAI is simply priced per server, with no additional data storage fees.

SMARTDriveAI is a purpose-built comprehensive solution targeted at the critical need for monitoring of drives (SSD, HDD, NVMe SSD) within servers.

You said you've trained your ML models on over 500,000 drives. What did you learn?

Our ML models and advanced analytics are trained on real world drive health and performance data. We've learned that a majority of drive failures can be predicted based on drive health and performance data. We've created ML models because some of the drive failures had nuanced changes in the health and performance data, changes that were hard to observe. We also learned that some errors did not require the use of ML models to detect them, analytics were sufficient.

SMARTDriveAI has dozens of known drive error signatures from numerous vendor drives encoded in its ML models and advanced analytics, all continuously updated with real time drive failure behavior and capable of applying it at scale.

Which servers do SMARTDriveAI support?

SMARTDriveAI currently supports both Windows and Linux based servers. Current OS's supporting include:

Linux:
- Amazon 3 and above
- Centos 07 and above
- Debian 10 and above
- Fedora 29 and above
- Opensuse 15 and above
- Redhat 7 and above
- Sles 15 and above
- Ubuntu 18 and above

Windows:
- Windows 10 and above
- Windows server 2012 and above

MAC:
Coming soon!

Contact sales@jedaanalytics.com for any operating system not listed.

What actions should I perform when I get SMARTDriveAI alerts?

The actions taken are dependent on the applications using the drive, their critically and the physical environment. For example, whether the drive is running mission critical applications, or whether the drive is in a RAID set and what the RAID rebuild times may be. The general process should be to migrate critical applications away from the drive predicted to fail. This could mean to optionally stop the server or take the drive offline and either reformat the drive, depending on age and history and restore the drive or just retire and replace the drive. The earlier the warning of an impending drive failure, the more convenient for IT operations, who can simply schedule any actions deemed appropriate.

Where do I find the correct server data collector?

The collectors can be found by selecting the 'Collector Downloads' link on your smartdriveai.com profile page.

How do I install a data collector?

General download instructions are included in the response email sent by SMARTDriveAI after user registration. Additionally, after downloading the collector and uncompressing, detailed installation instructions (README), along with the required files, can be found in the company collector directory. If you run into any issues contact support@jedaanalytics.com, they will be happy to assist.

Can I use SMARTDriveAI if my Data Center is not connected to the internet?

SMARTDriveAI requires only drive health and performance data along with any specific status to be sent to the cloud, never will anything be sent to your servers, so no special firewall rules are needed. However, if that is a problem, we have custom solutions for air gapped Data Centers. Contact sales@jedaanalytics.com for more details.

What if my servers have more than 32 drives?

Not a problem, SMARTDriveAI supports any number of drives per server. Contact sales@jedaanalytics.com for quotes for servers supporting greater than 32 drives.

If I already collect SMART drive data from my servers, can I still use SMARTDriveAI?

Yes, SMARTDriveAI can work in conjunction with other methods of SMART drive data collection. In Linux systems, we also work seamlessly with the widely used open-source Collectd data collector, or as a stand-alone collector. Although SMARTDriveAI collects some important additional non-SMART system data, we can offer solutions where your current SMART collection data can be used with a SMARTDriveAI solution. Contact sales@jedaanalytics.com for those environments.