We received excellent feedback on the new diagnostics and analytics capabilities of Visual Studio Application Insight announced at //Build 2016.
Here are couple customer success stories in their own words.
Quorum
Quorum (qbsol.com) makes innovative software for hydrocarbon and energy business management. We’re advancing the industry with a platform of integrated solutions proven to drive customer profit throughout the energy value chain. Quorum’s cloud-based software for production operations and field operations runs on Microsoft Azure. This means oil and gas companies benefit from unparalleled speed and performance.
The new analytics capability added to App Insights provides our team with immediate access to critical application and service health telemetry. I see this as being the go-to tool when we are trying to answer questions about what is happening that last, five, 30, or 90 minutes by writing a SQL-like query on our app data and get an immediate response. Now, when we want to see trends in data over time, proactively making changes to improve the platform and its services, we need something like Power BI.
For any tool to be useful, it has to have as few barriers between the decision to use it and the value provided by it. Even with a single managed platform for data collection and storage, the real value is not achieved until the data can be analyzed, turning it into information that can be used. Little things like Intellisense I see as a differentiator between App Insights and other competitive services on the market.
Ultimately, I was looking for a service to leverage, not another cluster to manage. We initially were working with an open source search and analytics engine and looking to move all of our telemetry to it. After a few occurrences of cluster unavailability, we needed to change direction and find an alternative that required less overhead. We chose App Insights analytics for the solution.
The long-term potential for new analytics features in App Insights exists both in the cloud and on-premises, desktops and smart devices, service health and performance, and device and sensor data throughout the oil and gas landscape. – Shawn Cutter, vice president, Quorum Business Solutions
OSIsoft
OSIsoft, a 30-year old software company based in the San Francisco Bay Area, is a global leader in industrial process data and operational intelligence. OSIsoft’s PI System is used in over 65% of the fortune 500 industrial companies to gather, store, process and present large volumes of sensor-based operations data for real-time decision support and advanced analytics.
At OSIsoft, we started developing a cloud-native enhancement of our sensor-based data infrastructure technology and very soon we were on the lookout for a telemetry solution that would work with a distributed, large-scale system. We needed something that would satisfy our twofold requirement:
- Ability to store large amounts of log and metric data at scale
- Ability to reliably query this data
Having explored multiple technologies that did not satisfy one or the other requirement, thanks to the close relationship we have with Microsoft, we finally ended up with Kusto, an internal Microsoft technology at that time. Once we started integrating Kusto into the telemetry framework, we were quite impressed with the data consumption rates. Kusto has allowed us to log telemetry data from multiple microservices running in a distributed framework, while ensuring mechanisms to adequately adapt for future changes via both schema and schema-less storage. “Having a highly scalable microservice architecture creates a hard challenge around telemetry. Kusto provided us with an ideal combination of scale, robustness and flexibility,” said Sai Roop Tetali, software developer at OSIsoft.
The SQL-like querying capabilities in Kusto are as impressive as its storage abilities. We are able to write rich queries that sift through vast amounts of data and quickly provide the information we need. Since we moved to Kusto we have seen a decided improvement in diagnosing issues with rapid turnaround. “Kusto provided us with extremely good querying capabilities. Having a SQL like querying syntax made a huge difference in the ramp up time required for the team to quickly get onboard and immediately start getting much deeper insight into how our services operated at scale,” said Tetali.
We are also able to track usage of our distributed application across multiple tenants by querying the raw telemetry data we store in Kusto.”In the past we’ve been able to view static data. Although helpful, it doesn’t really tell a story,” said Allison Muraski, service operations product manager at OSIsoft. “Kusto tackles the difficult task of not only returning static metrics, providing aggregation and calculation, but rendering them in a way that you can answer a question quickly and without manual manipulation of the data.”
“Kusto is a huge leap from Application Insights. Kusto Explorer, especially, is a joy to use. Now that we count on easy consumption, we can plan to log more data about our application,” said Chad Chisholm, cloud services initiative lead at OSIsoft. Thanks to Kusto not only are we able to store large amounts of high frequency telemetry data, we also have rich querying capability that has allowed us to develop insights into where and how our application is functioning under various conditions. The feedback from the team using Kusto has been so positive, we are now rolling it out to multiple other teams at OSIsoft. – Manas Talukdar, software engineering team leader, OSIsoft, LLC
Note: OSIsoft mentions the internal code name “Kusto” for Application Insights Analytics capability as referenced in Brian Harry’s blog.
Check out these videos on Channel 9 that go into detail on Application Insights Analytics features for more information.