gbrodmanandGitHub 6181648af6 Implement more efficient bulk put/update in JPATM (#3237)
This is more complicated than the bulk load unfortunately.

Previously, for update/put or their plural equivalents, we'd have to run
one or two SELECT statements per entity to check if it exists already
(and for Hibernate to load the object into the context to merge it with
the input).

Both updateAll and putAll now run, per input class, one large SELECT statement
to find out which entities already exist, which also loads the entities
into the persistence context (which is basically a cache). For updateAll,
we require that all of the entities already exist. For putAll, we have a
split path
- for entities that already exist, we just call a normal merge()
- for entities that don't already exist, we run merge() while instructing
Hibernate that the object is transient (i.e. new). This means that we
get the "put" behavior (not modifying the original object) while avoid a
superfluous SELECT statement that'd check again to see if the object exists in
the DB already.

For new objects in putAll, we also use EntityManager::merge (rather than
insert/persist) so that Hibernate creates a deep copy of the entity
instead of mutating the caller's instance in place. To prevent
Hibernate's DefaultMergeEventListener from firing a redundant SELECT
statement when merging a known-new entity with a non-null ID,
TransactionInfo implements Interceptor and overrides
isTransient(Object) to return Boolean.TRUE while merging a
known-transient entity.

We add comments clarifying that "insert" will modify an entity in place
(i.e. UpdateAutoTimestamp or generated IDs) whereas "put" will not. This
has always been the case; the comments are just to clarify.
2026-10-20 17:06:21 +00:00
2026-09-28 19:35:02 +00:00
2026-03-30 17:15:55 +00:00
2022-12-02 22:28:33 -05:00
2024-10-04 22:46:25 +00:00
2026-03-30 17:15:55 +00:00
2026-03-30 17:15:55 +00:00
2026-03-30 17:15:55 +00:00
2019-07-15 17:49:22 -04:00
2017-05-23 17:22:49 -04:00

Nomulus

Internal Build FOSS Build License Code Search
Build Status for Google Registry internal build Build Status for the open source build License for this repo Link to Code Search

Nomulus logo

Overview

Nomulus is an open source, scalable, cloud-based service for operating top-level domains (TLDs). It is the authoritative source for the TLDs that it runs, meaning that it is responsible for tracking domain name ownership and handling registrations, renewals, availability checks, and RDAP requests. End-user registrants (i.e., people or companies that want to register a domain name) use an intermediate domain name registrar acting on their behalf to interact with the registry.

Nomulus runs on Google Kubernetes Engine and is written primarily in Java. It is the software that Google Registry uses to operate TLDs such as .google, .app, .how, .soy, and .みんな. It can run any number of TLDs in a single shared registry system using horizontal scaling. Its source code is publicly available in this repository under the Apache 2.0 free and open source license.

Getting started

The following resources provide information on getting the code and setting up a running system:

If you are thinking about running a production registry service using our platform, please drop by the user group and introduce yourself and your use case. To report issues or make contributions, use GitHub issues and pull requests.

Capabilities

Nomulus has the following capabilities:

  • Extensible Provisioning Protocol (EPP): An XML protocol that is the standard format for communication between registrars and registries. It includes operations for registering, renewing, checking, updating, and transferring domain names.
  • DNS interface: The registry provides a pluggable interface that can be implemented to handle different DNS providers. It includes a sample implementation using Google Cloud DNS, as well as an RFC 2136 compliant implementation that works with BIND. If you are using Google Cloud DNS, you may need to understand its capabilities and provide your own multi-AS solution.
  • Registration Data Access Protocol (RDAP): A JSON API that returns structured, machine-readable information about domain name ownership. It is essentially a newer version of WHOIS.
  • Registry Data Escrow (RDE): A daily export of all ownership information for a TLD to a third party escrow provider to allow take-over by another registry operator in the event of serious failure. This is required by ICANN for all new gTLDs.
  • Premium pricing: Communicates prices for premium domain names (i.e., those that are highly desirable) and supports configurable premium registration and renewal prices. An extensible interface allows fully programmatic pricing.
  • Billing history: A full history of all billable events is recorded, suitable for ingestion into an invoicing system.
  • Registration periods: Qualified Launch Partner, Sunrise, Landrush, Claims, and General Availability periods of the standard gTLD lifecycle are all supported.
  • Brand protection for trademark holders (via TMCH): Allows rights-holders to protect their brands by blocking registration of domains using their trademark. This is required by ICANN for all new gTLDs.
  • Registrar support console: A self-service web console that registrars can use to manage their accounts in the registry system.
  • Reporting: Support for required external reporting (such as ICANN monthly registry reports, CZDS, Billing and Registration Activity) as well as internal reporting using BigQuery.
  • Administrative tool: Performs the full range of administrative tasks needed to manage a running registry system, including creating and configuring new TLDs.
  • Secure storage of cryptographic keys: A keyring interface is provided for plugging in your own implementation (see configuration doc for details), and an implementation based on Google Cloud Secret Manager is available.
  • TCP Proxy: Nomulus is built on top of the Jetty container that implements the Jakarta Servlet specification and only serves HTTP/S traffic. A proxy to translate raw TCP traffic (e.g., EPP) to and from HTTP is provided. Instructions on setting up the proxy are available. The proxy can either run in a separate cluster and communicate to Nomulus public HTTP endpoints via the Internet, or as a sidecar with the Nomulus image in the same pod and communicate to it via loopback.

Additional components

Registry operators interested in deploying Nomulus will likely require some additional components that need to be configured separately.

  • A way to invoice registrars for domain name registrations and accept payments. Nomulus records the information required to generate invoices in billing events.

  • Fully automated reporting to meet ICANN's requirements for gTLDs. Nomulus includes substantial reporting functionality, but some additional work will be required by the operator in this area.

  • System status and uptime monitoring.

Outside references

S
Description
No description provided
Readme Apache-2.0
157 MiB
Languages
Java 88.2%
HTML 8.7%
TypeScript 1.6%
Python 0.5%
Shell 0.5%
Other 0.2%