---
title: "Clustering"
description: "3 posts about Clustering, summarised, each linking to the original."
---

# Clustering
> 3 posts about Clustering, summarised, each linking to the original.

## Articles

### [Understanding Private Evolution as Learning-Augmented Clustering](https://yomu.fyi/post/understanding-private-evolution-as-learning-augmented-clustering.md)
- Company: [Apple](https://yomu.fyi/company/apple.md)
- Published: Nov 10, 2026
- Type: Problem & solution

Private Evolution (PE) is a differentially private algorithm for synthetic data generation that performs much better in practice than worst-case Wasserstein analyses would predict. The authors recast PE as generative model-augmented Wasserstein learning and show theoretically that when taking into account the use of a generative model that is able to capture something about the true distribution, much better performance bounds can be obtained. For example, if the generator gives samples in the same low-dimensional space as the distribution, then sample complexity depends on intrinsic, not ambient, dimension. They also show that standard variants of PE can fail to converge on simple well-clustered instances, and propose a new geometry-aware version of PE with provable convergence on such instances. Experimentally, the new algorithm is competitive with standard baselines and can improve recall.


### [JBoss EAP clustering in OpenShift: KUBE\_PING discovery](https://yomu.fyi/post/jboss-eap-clustering-in-openshift-kube-ping-discovery.md)
- Company: [Red Hat](https://yomu.fyi/company/red-hat.md)
- Author: Francisco De Melo Junior, Alexander Barbosa Ayala
- Published: Nov 3, 2026
- Type: Tutorial

This article explores the JGroups KUBE\_PING discovery protocol for clustering JBoss EAP 7 and JBoss EAP 8 in Red Hat OpenShift 4. KUBE\_PING queries the Kubernetes API directly for pod IP addresses, then filters them by namespace and labels to establish peer-to-peer TCP connections. Configuring KUBE\_PING requires setting three environment variables: JGROUPS\_PING\_PROTOCOL, KUBERNETES\_NAMESPACE, and KUBERNETES\_LABELS. The default service account must be granted the view role on the namespace to avoid HTTP 403 errors. In JBoss EAP 8, clustering is not enabled by default and must be provisioned via Galleon layers during the build process.


### [Cross-store product clustering across millions of stores (2026) - Shopify](https://yomu.fyi/post/cross-store-product-clustering-across-millions-of-stores-2026-shopify.md)
- Company: [Shopify](https://yomu.fyi/company/shopify.md)
- Author: Ilia Kuchumov
- Published: Oct 20, 2026
- Type: Problem & solution

Shopify’s cross-store product clustering system groups listings of the same product across different merchants into a single global Universal Product Identifier. The approach avoids comparing every pair by generating candidate pairs using approximate nearest-neighbor search across billions of vector embeddings. Scoring uses three models of increasing cost: L1 is a fast GBDT classifier, L2 is a fine-tuned multilingual cross-encoder that reads both listings together, and L3 adds images. The UPI builder evaluates proposed merges using a trained model that weighs evidence, handling non-transitive pairwise judgments. The system caches scores and uses priority queues for expensive inference, enabling daily cluster updates while background GPU inference continues refining matches.
