--- title: Postgres 18's UUIDv7 shrinks indexes, AWS Glue 6.0 gets... description: Postgres 18 keeps stacking small wins today: native uuidv7() cuts index bloat by 25% versus UUIDv4, and EXPLAIN ANALYZE now includes buffer stats by default.... canonical: https://daily.dev/posts/postgres-18-s-uuidv7-shrinks-indexes-aws-glue-6-0-gets-30-cheaper-zuz5qqju2 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Postgres 18's UUIDv7 shrinks indexes, AWS Glue 6.0 gets 30% cheaper | daily.dev og:description: Postgres 18 keeps stacking small wins today: native uuidv7() cuts index bloat by 25% versus UUIDv4, and EXPLAIN ANALYZE now includes buffer stats by default.... og:url: https://daily.dev/posts/postgres-18-s-uuidv7-shrinks-indexes-aws-glue-6-0-gets-30-cheaper-zuz5qqju2 og:image: https://api.daily.dev/og/posts/zUZ5Qqju2.png og:image:alt: Postgres 18's UUIDv7 shrinks indexes, AWS Glue 6.0 gets 30% cheaper og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Postgres 18's UUIDv7 shrinks indexes, AWS Glue 6.0 gets 30% cheaper **[Databases Digest](https://daily.dev/sources/databases_digest)** · 4 min read · 5 upvotes · 0 comments ## Summary Postgres 18 keeps stacking small wins today: native uuidv7() cuts index bloat by 25% versus UUIDv4, and EXPLAIN ANALYZE now includes buffer stats by default. AWS Glue 6.0 ships with a 30% price cut and full Iceberg v3 support, landing in two separate posts today. A pair of connection pool benchmarks independently confirm that bigger isn't better - throughput peaks around 48 connections on a 4-core box and then falls off a cliff. MySQL shops running Galera Cluster have until September 30 to figure out their migration path before the patch pipeline dries up. ## Content **TLDR:** Postgres 18 keeps stacking small wins today: native uuidv7() cuts index bloat by 25% versus UUIDv4, and EXPLAIN ANALYZE now includes buffer stats by default. AWS Glue 6.0 ships with a 30% price cut and full Iceberg v3 support, landing in two separate posts today. A pair of connection pool benchmarks independently confirm that bigger isn't better - throughput peaks around 48 connections on a 4-core box and then falls off a cliff. MySQL shops running Galera Cluster have until September 30 to figure out their migration path before the patch pipeline dries up. --- ## Postgres 18's uuidv7() produces smaller, less fragmented indexes than UUIDv4 Benchmarks across a million rows show UUIDv7 primary key indexes running 30MB versus 38MB for v4, with 90% leaf density and zero fragmentation compared to v4's 71% density and roughly 50% fragmentation. The reason is boring but important: v7's embedded millisecond timestamp keeps inserts sorted at the right edge of the B-tree instead of scattering them randomly. Trade-offs still apply - v7 leaks creation time and can cause write contention on the rightmost page in heavily sharded setups, which is where a BIGINT-based alternative like pgEdge's snowflake extension comes in, though that has its own fragmentation ceiling around 100k inserts/sec cluster-wide. [Read more](https://daily.dev/posts/X5yVsDPZm) ## Two connection pool benchmarks agree: 48 beats 400 on 4 cores Two independent writeups sweeping PostgreSQL 17.11 from 1 to 400 connections found throughput peaking at 48 connections (18,039 tps) and dropping 37% by 400, with latency multiplying 13x. The popular (cores×2)+spindles formula predicts just 9 connections, way under the measured optimum, because it ignores time spent waiting on WAL, network, or locks. Both pieces land on the same diagnostic: stop counting connections and start watching queue depth - cl_waiting in PgBouncer, pending-threads in HikariCP - since that's what actually explains why Lambda functions blow through RDS connection limits even with pooling. [Read more](https://daily.dev/feed-by-ids?id=mVlgrYPkO&id=F0OqiZyNP) ## AWS Glue 6.0 goes GA with 30% lower pricing and Iceberg v3 AWS Glue 6.0 is generally available with a 30% price cut and full Apache Iceberg v3 support built on Iceberg 1.11.0, including a VARIANT type with shredding for semi-structured data, geometry/geography types, and deletion vectors for row-level updates. The runtime moves to Apache Spark 4.1 and Scala 2.13, adding Spark Declarative Pipelines and a real-time streaming mode with single-digit millisecond latency. No API changes are required - existing jobs pick up 6.0 via the --glue-version parameter or get auto-upgraded through the Spark Upgrade Agent. [Read more](https://daily.dev/feed-by-ids?id=Dkp07ZvdA&id=lEaQ7LqLv) ## PITR recovers a wiped production database in 0.63 seconds A demo shows what happens when Laravel's migrate:fresh --force wipes a production-sized Neon Postgres database (5,000 customers, 25,000 orders): the wipe takes 21 seconds, but point-in-time restore via a single Neon API call takes 0.63 seconds with no connection string change or redeploy. That makes nightly pg_dump backups look almost quaint by comparison for recovery point objectives. Suggested guardrails include Laravel's DB::prohibitDestructiveCommands(), separating migration credentials from DROP rights, and actually drilling the restore before you need it. [Read more](https://daily.dev/posts/eFprr3ZjK) --- ## Also notable - **MySQL Galera Cluster maintenance ends September 30:** MariaDB stops shipping new MySQL-wsrep builds after September 30, 2026, and even upgrading to MySQL 8.4 doesn't dodge it since wsrep builds stop at 8.4 while Galera's own maintenance ends the same date. [Read more](https://daily.dev/posts/m8WwiYmbY) - **DuckDB Java driver adds chunked columnar result reading:** DuckDB Java driver 1.5.3.0 introduces a DuckDBChunkedResult API that reads native columnar batches of 2,048 rows directly, skipping the per-value JNI overhead of the standard JDBC ResultSet for large result sets. [Read more](https://daily.dev/posts/o8w3naGZN) - **Composite index column order turns a 16.6ms scan into 0.04ms:** A walkthrough on Postgres 18 with 1 million seeded comments shows that putting equality columns first in a composite index, following the leftmost prefix rule, cuts a dashboard query with a LATERAL join from 436ms to 0.5ms. [Read more](https://daily.dev/posts/exxMe3LqV) - **MongoDB many-to-many refactor cuts query time from 4.7s to 2.9s:** Replacing an intermediate associative collection with an embedded array of device IDs eliminates a $lookup stage, dropping total elapsed time for 300 iterations from 105 to 62.5 seconds, still short of the sub-second target. [Read more](https://daily.dev/posts/UiixfWluy) - **Postgres 18 shows buffer stats in EXPLAIN ANALYZE without the BUFFERS flag:** PostgreSQL 18 includes buffer usage information in EXPLAIN ANALYZE output by default, so the explicit BUFFERS option is no longer needed when analyzing query plans. [Read more](https://daily.dev/posts/3X5LbpfD8) ## Similar posts on daily.dev - [PostgreSQL 18: 23x Faster Inserts With UUID V7](https://daily.dev/posts/postgresql-18-23x-faster-inserts-with-uuid-v7-nc2q5uwx2) · Planet PostgreSQL · 19 upvotes · 1 comments - [The Time Traveler's Primary Key](https://daily.dev/posts/the-time-traveler-s-primary-key-x5yvsdpzm) · Planet PostgreSQL · 0 upvotes · 0 comments - [Exploring PostgreSQL 18's new UUIDv7 support](https://daily.dev/posts/exploring-postgresql-18-s-new-uuidv7-support-uspoubwqo) · Hacker News · 0 upvotes · 0 comments --- Tags: [#aws](https://daily.dev/tags/aws), [#database](https://daily.dev/tags/database), [#postgresql](https://daily.dev/tags/postgresql) [View this post on daily.dev](https://daily.dev/posts/postgres-18-s-uuidv7-shrinks-indexes-aws-glue-6-0-gets-30-cheaper-zuz5qqju2) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"DiscussionForumPosting","mainEntityOfPage":"https://daily.dev/posts/postgres-18-s-uuidv7-shrinks-indexes-aws-glue-6-0-gets-30-cheaper-zuz5qqju2","headline":"Postgres 18's UUIDv7 shrinks indexes, AWS Glue 6.0 gets 30% cheaper","text":"Postgres 18 keeps stacking small wins today: native uuidv7() cuts index bloat by 25% versus UUIDv4, and EXPLAIN ANALYZE now includes buffer stats by default. AWS Glue 6.0 ships with a 30% price cut and full Iceberg v3 support, landing in two separate posts today. A pair of connection pool benchmarks independently confirm that bigger isn't better - throughput peaks around 48 connections on a 4-core box and then falls off a cliff. MySQL shops running Galera Cluster have until September 30 to figure out their migration path before the patch pipeline dries up.","url":"https://daily.dev/posts/postgres-18-s-uuidv7-shrinks-indexes-aws-glue-6-0-gets-30-cheaper-zuz5qqju2","datePublished":"2026-08-22T04:18:38.781Z","dateModified":"2026-09-13T21:34:28.612Z","author":{"@type":"Organization","name":"Databases Digest","logo":"https://media.daily.dev/image/upload/s--ir1hdV8b--/f_auto,q_auto/v1773839400/logos/databases_digest?_a=BAMAMiiu0","url":"https://daily.dev/sources/databases_digest"},"interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":5},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"isPartOf":{"@type":"WebPage","url":"https://daily.dev/sources/databases_digest","name":"Databases Digest"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Databases Digest","item":"https://daily.dev/sources/databases_digest"},{"@type":"ListItem","position":3,"name":"Postgres 18's UUIDv7 shrinks indexes, AWS Glue 6.0 gets 30% cheaper"}]} ```