Dedistributed PTR is the fastest way for cloud-based applications to obtain the published hostnames for IP addresses (the values of DNS PTR records). It is offered through the AWS Marketplace as a metered service and scales to any scale. Use it to enable your applications to perform billions of lookups daily with orders of magnitude less latency, reliably and at low cost.
For any IP address, obtain the corresponding hostname (PTR record value). Enrich your records.
Low latency in practice means 0.3 milliseconds of round trip time to retrive the PTR value for a single IP address. Each additional IP address included in a query adds a marginal latency of less than 400 nanoseconds, supporting batching. Up to 10,000 PTR records can be retrieved in a single API call. See benchmarks.
Use Dedistributed PTR as a drop-in replacement for a DNS query (or API call you would otherwise make to a self-maintained caching layer). Engineers add a small dependency to their builds, instantiate a client instance at the beginning of an application's lifecycle, and then call as needed while processing records.
Pricing is volume-based and billed hourly. We offer a Free Tier to support engineers and teams who want to try the service at no cost (an AWS account is required). At a price point of more than 100,000,000 lookups per US dollar, our commodity pricing is highly competitive even to self-managed solutions at scale.
We update the dataset weekly. The DNS PTR record values obtained from the Internet's public DNS are generally static for long periods, but a small number of them do change over time. Access to updated values is automatic.
With just six active sessions (and ample bandwidth), a streaming or batch application can perform 1 trillion lookups per day. By using Dedistributed PTR, you free up engineering resources to focus on what is important.
Dedistributed PTR is designed for applications which require hostname lookups to complete with consistent low latency, and or which process volumes in excess of 100 million per day. Generally this means streaming event and log processing applications, IoT-related applications, and systems performing security and threat analysis at scale.
The technology stack of these applications can be home grown, but more often than not they rely on open source and commercial infrastructure such as:
Latency is minimized when applications operate in the same Availability Zone as our endpoints, making Dedistributed PTR ideal for cloud-based workloads. However, on-premises use cases are supported too. Batch mode or latency-tolerant applications executed in on-prem environments can take advantage of the service's pricing, fixed latency, and the no-curves path to scale.
How does it work? We curate and maintain the entire 50 GB dataset in RAM on AWS-resident servers. A purpose-specific index, per AZ client-to-server matching, and a binary TCP protocol enable fast retrieval of any (or all) of the 1.3 billion public hostname records. There is never a cache miss.
The SDK supports retrival requests with up to 10,000 IP addresses per invocation. Multiple calls can be interleaved in the thread-safe versions of the client for multiple threads-per-VM pipelining.
See Developer Resources for more detail.
Pricing is tiered and by volume. We have a Free Tier. See pricing.
Dedistributed PTR is offered through the AWS Marketplace as a SAAS offering. Enrollment (setting up your subscription) is easy and billing is handled through the marketplace.
Fees associated with your organization's use of Dedistributed PTR are added to your monthly AWS services bill as a line item.
Enroll and try it out today! Visit the Dedistributed PTR service profile in the marketplace.