Innova logoCase study

Innova migrated to QuestDB

Innova chose QuestDB as part of their big data solution, which requires writing millions of records while querying a constantly changing data set for real-time analytics.

Innova background

Dollar iconLower operational costs

Workflow iconSimple setup and maintenance using QuestDB Docker image

Leaf iconIntegrations with existing tools using the REST API

Gauge iconReal-time results despite massive throughput

Voice iconReactive support from QuestDB engineering

Time iconReliable and fast query times for insights

Innova develops big data solutions for financial transactions, BI systems, IT infrastructure, security, and network operators. Our services include real-time analytics of the network infrastructure of the largest Telecommunications provider in Turkey.

In this case study, we summarize why Innova chose QuestDB, their migration experience, and the improvements they gained in query speed, maintainability, and compatibility.

Why Innova migrated to QuestDB

Innova found QuestDB because of its requirements to store and analyze massive amounts of data that needs to be served to their customers quickly. The data Innova collects needs to be presented to customers so that it’s easy to understand changes over time. The search for a database that can display this kind of information in a timeline as fast as possible led to QuestDB.

Innova’s client is one of the largest telecommunications providers in Turkey, serving internet connections to over 5 million customers. Infrastructure at this scale uses massive amounts of resources and generates a tremendous volume of data. Innova uses operational data from this provider to show their customers information about their own internet connections, such as the quality of service, download speed, upload speed, bandwidth, and more.

How Innova collect and analyze big data with QuestDB

Innova store the bandwidth data of devices in Fiber To The Home (FTTH) topology from collectors, which run on hourly intervals. These collector jobs contain time series for each device in JSON format and based on this metadata, Innova produce visualizations for their customers to have insights into their network quality. For a customer, this means that they have access to charts, tables, and line graphs so they can see the state of their connectivity for the last 30 days.

Chart showing the average transaction duration for QuestDB on a given day

Why Innova use a time series database for big data

The original stack built by Innova relied on MongoDB to store connectivity information, which was convenient at first. However, it soon became apparent that MongoDB was tightly dependent on physical resources and required more powerful hardware that they were happy with allocating. Moreover, as data sets grew in MongoDB, query speeds degraded below what was acceptable for their customers.

For communicating with QuestDB, Innova uses the QuestDB Docker image for running production instances and the REST API to query data for charts and insights. On average, Innova is writing hundreds of millions of records per day while performing calculations on an ever-changing data set. Usually, Innova will keep 30 days of data in QuestDB and delete older partitions when they become less useful for actionable insights.

QuestDB allows us to query data while writing millions of records. It is an excellent database for time series analysis and can efficiently store our data. QuestDB’s community is constantly growing and its popularity is on the rise.

Erdem Aydemir, Innova