Frequently Asked Questions

OmniSci will not refund money for a billing cycle already underway. If you decide to cancel your instance of OmniSci Cloud, we will trigger your plan to expire at the end of the current cycle. That also means that you’ll have access to your instance until that cancellation date.

If you upgrade your plan, those changes will take effect immediately, and your billing will adjust to the rate of your new plan.

If you downgrade your plan, those changes will take effect at the end of your billing cycle. Your rate will adjust to the lower amount at the start of your next billing cycle.

OmniSci will provision your OmniSci Cloud instance on either Amazon Web Services or Google Cloud Platform, at our discretion. All data will be hosted within the United States.

No. We’ve included those datasets as a courtesy, so you can get started immediately. Only additional datasets that you import count against that limit. You can import some of those other publicly available datasets from this S3 repository or bring in datasets that you already have available.

There are no limits on the number of columns. However, a really wide data set may potentially result in an error. If you receive that error, you will need to reduce the size of your dataset.

After an idle period, all instances of OmniSci Cloud that are not in active use are put to sleep. Once you come back to use your instance, you may experience a brief delay as your instance of the OmniSci Cloud analytics platform wakes back up. The OmniSci Cloud X offering does include an “always on” option.

OmniSci uses a number of leading security techniques to secure your data. All customer data in OmniSci Cloud is encrypted at rest, and all communication between you and the OmniSci Cloud servers is encrypted along the way using industry-standard encryption standards. OmniSci Cloud has multiple security controls built in at every level to protect against malicious use, unauthorized access, data loss, and downtime of systems. Finally, OmniSci provisions hardware with leading cloud platform providers such as Amazon AWS and Google, that maintain industry-leading security protocols.

OmniSci may collect metadata and statistics about customer system usage in order to improve the service, and we only access data for support purposes. Please see our Privacy Policy and Terms and Service for more information.

You can cancel your OmniSci Cloud subscription at any time. When you confirm that cancellation decision you will lose access to that instance. Your data will be retained for a period of time in case you decide to reactivate your subscription, after which time it will be permanently deleted from the system. If you wish to delete your data outside of the standard procedures, please contact OmniSci support.

Your data will remain securely in OmniSci Cloud for 30 days following your trial expiry. At any time in that period you can activate a paid subscription and your data and dashboards will convert over.

When your trial ends, OmniSci will retain your data for a period of time in case you wish to become a subscriber. After that time it will be permanently deleted from the system. If you wish to have the data deleted outside of the automated process, or you wish to cancel your trial prior to the expiry date, please contact OmniSci support.

Certainly not. You can sign up for a subscription to either Base, Standard 1, Standard 2, or Standard 12 for OmniSci Cloud at any time during your trial. However, billing for that subscription will not begin until the day after your 14 day trial period is over. If you choose Cloud X as your cloud based analytics platform, you can contact us at any time to discuss a managed analytics cloud service.

Each plan specifies the allotted GPU RAM, CPU RAM, and Storage for that plan.

As an in-memory database, the amount of GPU RAM is what limits the amount of data that you will be able to explore at any one time in OmniSci. The actual number of records you can explore in dashboards varies based on the type of data, the number of columns being included in the query, and how it is encoded on import to the OmniSci database. 30,000,000 rows per 2GB of GPU RAM is a reasonable approximate, noting that we have seen some customers fit in excess of 50,000,000 rows per 2GB of GPU RAM in OmniSci Cloud. For example, a 12 GB GPU allotment roughly translates to approximately 180,000,000 rows (although this can be much higher depending on your data and how it is encoded).

OmniSci employs various methods to reduce the amount of GPU RAM required and the amount of will be less than the uncompressed source (except in very rate cases). For example, OmniSci is a columnar database, so only the columns required for a query will be loaded into memory. This means that a 50 column CSV file, which only needs to report 10 columns across charts in a dashboard, will only have those 10 columns loaded into memory. As another example, for low cardinality text fields (many of the fields contain words that are contained in other rows in the same column), the field can be dictionary encoded which means that each unique word in a field is assigned a number, and then those number-word pairs are stored in a single “dictionary” for lookup. This means a word repeated many times in that column takes up a fraction of the space it might otherwise, and therefore reduces the amount of data in GPU memory. For additional information about the design of tables to maximize performance, please see our documentation.

Some limits in the system are hard limits and some are soft. Hard limits will return an error when the limit is met, and soft limits will result in a notification.

Team functionality is included in every Cloud package.

There are five categories of users in OmniSci Cloud: Super Admin, Admin, Developer, Editor, and Viewer.

No, each registered email address can only be assigned to one Team.

Each plan comes with a write user limit and a total user limit. See below:

Both the Super Admin and Admin can: assign the Developer, Editor, or Viewer roles to individuals; add / update / deactivate users; upload and manage datasets; and edit and view dashboards. Only the Super Admin can assign the Admin role to users and manage system settings.

Both Admins and Editors can upload and manage datasets, edit and view dashboards. Only Admins can edit user roles and add / update / deactivate users.

Dashboards in OmniSci are shared with all users automatically.

All tables are shared automatically with all users. Only the Editor, Developer, Admin, and Super Admin have access to the Data Manager and SQL Editor to view the table structure and perform queries.

Yes, the user would be assigned a Viewer role and would see only the Dashboards menu, and all the dashboards are read-only for that user.

Yes, Teams is available during the 14 day trial period.

The team is created by automatically when the first user registers, and that user becomes the Super Admin. Go to Settings > Team > Invite Users to add team members.

A Super Admin or an Admin can deactivate a user at any time by going to Settings > Team > Edit. Select the user who is no longer on your team, scroll down and select Deactivate User.

Read more about OmniSci Cloud plans here. Each plan will specify if it includes the API.

The OmniSci Cloud API is compatible with the following existing OmniSci connectors, each of which have instructions on getting started:

• mapd-connector
• PyMapD
• JDBC (Available in the core package, sample code can be obtained here)

The OmniSci API is based on Apache Thrift and instructions on how to generate Thrift bindings are referenced in our documentation. For users who wish to generate client-side interfaces for various programming languages.

Super Admins, Admins, and Developers can each create a single read key and a single write key.

API request limits are governed by the OmniSci Cloud Fair Use Policy.

The IP whitelisting feature allows you to specify which source IP addresses are allowed to connect to your OmniSci instance via the API.

You are allowed to enter a maximum of 6 subnets or IP addresses.

Yes, but you must enter at least 1 subnet or IP address if you want to enable the IP whitelisting feature. If you specify no subnets then all IPs will have access.

The changes will be reflected immediately in the web front end, and may take up to five minutes to be propagated throughout the infrastructure.

IP Session Lock is one of the default security controls that are in place to protect from an unauthorized user from abusing your service. It means that the session is locked to the IP address from which that session initially authenticates. If a user manages to steal the session ID and attempt to connect from a different IP address than the one that created the session then the connection will be dropped. There are some edge cases where applications may require the ability to change their source IP address within the same session. We recommend that this is the only time that this feature be disabled.

OmniSci does not record or store API secrets in any way. Once an API secret is issued and lost, it is not recoverable. You can request a new API secret through the Developer tab in the settings menu of the OmniSci Cloud portal.

OmniSci strives to offer a package to suit all of our user’s needs. If you need a customized offering, contact us at cloud@omnisci.com.