OffNet Newsroom

Daily topic roundup

AWS

Tuesday, October 13, 2026 · 7 stories, curated & summarized — click any story for the source.

AWS has made G7 instances available in Europe (Frankfurt) and Europe (Stockholm), featuring NVIDIA RTX PRO 4500 Blackwell Server Edition GPUs. These instances offer up to 4.6x better AI inference performance and 2.1x improved graphics capabilities compared to the previous G6 generation. The hardware accelerates workloads including language models, video analysis, real-time rendering, and data analytics.

  • G7 instances are now live in EU-Frankfurt and EU-Stockholm regions for local latency needs.
  • Powered by NVIDIA RTX PRO 4500 Blackwell GPUs for enhanced AI and graphics throughput.
  • Delivers up to 4.6x AI inference and 2.1x graphics performance gains over G6 instances.
  • Ideal for deploying LLMs, video transcoding, and spatial computing workloads in Europe.
COMPARISONG7 vs G6 Performance GainsAI Inference4.6xGraphics2.1x

Amazon S3 Express One Zone is now live in four new regions: Europe (London), Asia Pacific (Jakarta and Hong Kong), and Africa (Cape Town). This single-AZ storage class offers consistent single-digit millisecond latency, up to 10x faster access than S3 Standard, and up to 80% lower request costs. The service supports latency-sensitive workloads like ML training and AI search caching, bringing the total available regions to 19.

  • S3 Express One Zone now covers London, Jakarta, Hong Kong, and Cape Town.
  • Delivers single-digit millisecond latency for latency-sensitive applications.
  • Performance is up to 10x faster than S3 Standard with 80% lower request costs.
  • Ideal for ML training, interactive analytics, and AI search engine caching.
  • Total global availability for this class has reached 19 AWS Regions.
BY THE NUMBERSS3 Express One Zone Growth19Total AWS Regions AvailableExpanding to London, Jakarta, Hong Kong, and Cape Town

Amazon OpenSearch Service now allows encrypting individual indexes at rest using distinct AWS KMS customer managed keys. This capability enables granular data protection for multi-tenant environments where different tenants require separate key ownership. Users can configure this by setting the encryption mode to Native during domain creation or via CLI, then registering additional keys to apply to specific indexes.

  • Encrypt each OpenSearch index with a unique AWS KMS customer managed key.
  • Ideal for multi-tenant architectures requiring strict data isolation and key ownership.
  • Enable by setting encryption mode to Native when creating or updating the domain.
  • Register multiple keys with the domain to assign them to specific indexes.
CHECKLISTSet Up Per-Index EncryptionSet encryption mode to NativeRegister multiple KMS keysAssign unique keys to indexesEnforce strict data isolation
Hacker News (100+ points) general

Intel D270 GPU AI Accelerator features up to 480 GB LPDDR5X memory

Intel has introduced the D270 discrete GPU designed for data center AI workloads. The accelerator is equipped with up to 480 GB of LPDDR5X memory to support high-capacity model inference and training. This release positions Intel to compete more directly in the specialized AI hardware market.

  • Intel enters the high-memory AI accelerator space with the D270 discrete GPU.
  • Up to 480 GB LPDDR5X capacity addresses large model memory bottlenecks.
  • Targeted at data center environments requiring significant parallel compute.
  • Signals continued investment in Intel's discrete GPU product line.

Ukrainian drone attacks on Yandex datacenters have caused significant outages for its cloud services and payment systems. The infrastructure damage has reduced network connectivity to approximately 30% of normal levels. Users are currently experiencing difficulties accessing funds and relying on workarounds by engineering teams.

  • Physical infrastructure attacks directly impact cloud availability and payment processing.
  • Network connectivity dropped to 30%, indicating severe localized or systemic disruption.
  • Engineering teams are actively implementing workarounds to restore critical services.
  • Reliance on single-region datacenters poses a critical risk for service continuity.
  • Geopolitical conflicts now directly threaten digital financial and cloud infrastructure.

Amazon GuardDuty expands RDS Protection to identify data exfiltration and destruction attacks against Amazon Aurora PostgreSQL and RDS for PostgreSQL. This capability leverages machine learning and threat intelligence to monitor query patterns for anomalies, requiring no agents or configuration changes. The update addresses threats where attackers use valid credentials to steal or wipe data after initial compromise.

  • Covers Aurora PostgreSQL and RDS for PostgreSQL specifically.
  • Detects exfiltration and destruction, not just login anomalies.
  • Agentless deployment with no infrastructure or config changes needed.
  • Uses ML-based anomaly detection and integrated threat intelligence.
  • Targets insiders or compromised accounts performing malicious queries.
CHECKLISTGuardDuty RDS Protection TakeawaysCovers Aurora and RDS for PostgreSQLDetects exfiltration and destructionAgentless with no config changesUses ML anomaly detectionTargets compromised account queries

AWS has updated Amazon EventBridge custom event buses to support centralized cross-account sharing with guaranteed ordered delivery. The update introduces event replay capabilities, content-based deduplication, and simplified subscriber management to streamline multi-account architectures. Additionally, a new publisher/subscriber-based pricing model replaces previous structures to align costs with usage patterns.

  • Cross-account event buses now guarantee message ordering, critical for stateful workflows.
  • Event replay allows reprocessing data without re-sending from publishers, aiding debugging.
  • Content-based deduplication reduces storage costs and prevents processing duplicates.
  • New pricing model shifts to publisher/subscriber rates, potentially lowering costs for high-volume setups.
  • Simplified subscriber management reduces operational overhead in large multi-account environments.
↑