The world's first public NTP service specifically housed in financial market data centers and optimized for algorithmic trading
SALT LAKE CITY, Nov. 7, 2023 /PRNewswire/ -- Databento is pleased to announce the recent launch of its public time service for algorithmic traders. The service is free and available to anyone over the internet and in the public cloud. Anyone can synchronize their clocks against the company's time service by pointing their NTP client at ntp.databento.com.
Why provide a free time service for algorithmic trading?
The company was inspired to provide a free service after many customers reported negative latencies when differencing local timestamps against outbound timestamps on messages from Databento's real-time data feeds, due to their unprecedented low latency over the internet.
Christina Qi, Databento CEO, said, "This was mostly an artifact of a time synchronization problem between the customers' clock source and Databento's. As our feeds got faster, the effect of time synchronization became more dominant and made it difficult to measure latency precisely. Many of our customers have used other market data feeds over public WAN, but they've never experienced this before. We wanted to provide a way for them, and anyone else, to get a more precise time sync against our internal clocks. This time service will enable algorithmic traders to take full advantage of the speed and timestamping accuracy of our data feeds."
How the Databento public time service works
Databento's time service is composed of a pool of stratum 1 NTP servers at the CyrusOne Aurora I and Equinix NY4 data centers, making it the world's first public NTP service that is specifically housed in financial data centers and optimized for systematic trading users.
The margin of error in time synchronization depends primarily on two factors: geographical distance and the number of network hops from the time source. Databento's time service helps users obtain more accurate timestamps than common time providers like Google and Cloudflare, which usually have time servers situated in general-purpose data centers that are situated farther from the popular colocation and proximity hosting sites used by financial firms.
Another important reason to use Databento's time service is the quality of its clock source. The time servers for its service are synchronized against the same PTP time source used for supplemental receive and outbound timestamps in the company's data, within 200 nanosecond tolerance. In turn, the company's PTP source is a pair of GPS-synchronized, high-precision grandmaster clocks with under 25 nanosecond root-mean-squared error against UTC. The servers themselves are configured to allow polling frequencies of nearly 10,000 per hour—much higher than most public time sources.
Who should use the Databento time service?
Databento's time service is ideal for users using proximity hosting services and public cloud providers like AWS, Azure, and Google Cloud, and for customers whose servers are located in the Greater Chicago and New York metro regions.
For users requiring the highest levels of accuracy, the company also offers a managed PTP service for its colocated customers. You can learn more about performance optimization on the company's docs site or contact sales.
About Databento
Databento's data-as-a-service (DaaS) platform makes it simpler and faster to access financial market data. Founded by traders and engineers from leading quantitative hedge funds and proprietary trading firms, Databento's self-service model allows users to instantly pick up real-time exchange feeds and access petabytes of historical market data. Users have the flexibility to only pay for what they use or opt for flat-rate pricing. Databento's servers are hosted in the colocation facilities of various trading venues for low-latency and high-fidelity data capture directly from the source.
To learn more about Databento, visit www.databento.com.
Media contact:
Jennifer DeRome
617-553-4000
[email protected]
SOURCE Databento
![](https://rt.prnewswire.com/rt.gif?NewsItemId=PH59862&Transmission_Id=202311070843PR_NEWS_USPR_____PH59862&DateId=20231107)
Share this article