NewApp Digest
search bot

architecture

Android Malware Caught by Teaching AI to Read Apps as Pictures
AppWizard
October 11, 2026

Android Malware Caught by Teaching AI to Read Apps as Pictures

A research team from Turkey, led by Erdal Başaran and Ömer Okucu of Ağrı İbrahim Çeçen University, along with Yusuf Alaca from Hitit University, has developed an innovative detection framework for Android malware that utilizes a visual approach. This framework transforms key features of Android applications into images, allowing a Vision Transformer to differentiate between benign software and malware. The researchers created two types of images from static and dynamic analysis features: a conventional 2D grayscale image and a QR code image. Separate Vision Transformer models were trained on each image type, and spatial pooling features were extracted from deeper layers to create a comprehensive representation of application behavior. To refine the fused feature vector, Recursive Feature Elimination (RFE) was employed, enhancing classification clarity and reducing computational load. The final classification utilized a majority-voting ensemble strategy, achieving a detection accuracy of 98.72 percent, surpassing standalone models. The study demonstrates that the multimodal system outperforms individual grayscale or QR code pathways. The dataset used is publicly available, promoting reproducibility. The framework's reliance on static and dynamic analysis features allows it to adapt to new threats, and the novel QR code representation offers distinctive visual signatures of malicious behavior. However, challenges remain regarding feature extraction quality and computational demands.
A practical guide to cost optimization with Lakebase Postgres
Tech Optimizer
October 11, 2026

A practical guide to cost optimization with Lakebase Postgres

Lakebase is a fully managed Postgres database designed for modern application development, featuring an architecture that separates storage and compute with a serverless compute layer. This design allows for cost efficiencies through various mechanisms: 1. **Branching**: Developers can create isolated environments for development, testing, or experimentation without duplicating storage costs, as branches share the same underlying storage. 2. **Autoscaling**: Lakebase adjusts compute resources based on activity levels, allowing users to pay only for the compute they use. It can scale down during low activity and can be suspended entirely after inactivity, reducing costs to zero. 3. **Read Replicas and High Availability**: The separation of storage and compute allows for adding read replicas and high availability without incurring additional storage costs, as these instances utilize the same storage layer. 4. **Synced Tables**: Integration with the Databricks Intelligence Platform enables efficient data syncing, allowing users to sync only the necessary working set of data, which helps avoid unnecessary storage and sync costs. 5. **Sync Modes**: There are three sync modes (Snapshot, Triggered, Continuous) for transferring data from the Lakehouse to Lakebase, each with different cost implications based on data freshness requirements. 6. **Right-sizing Compute**: Users can set the initial compute range during project provisioning to avoid over-provisioning and unnecessary costs. 7. **Monitoring and Metrics**: The Lakebase Metrics dashboard provides insights into working set size and cache utilization, helping users optimize their compute sizing and performance. 8. **Point-in-Time Restore (PITR) and Snapshots**: PITR maintains history for recovery within a configurable window, while snapshots offer discrete recovery points. Both are priced lower than standard storage, making them cost-effective options for data recovery. Lakebase costs are categorized into compute, storage, and serverless pipeline compute for syncing data. Compute is measured by CU usage, while storage includes branch storage, PITR history, and snapshot storage, all tracked separately. Users can access detailed usage and cost information through system billing tables.
Amazon Aurora PostgreSQL now supports direct querying of Apache Iceberg and Parquet data in your data lake
Tech Optimizer
October 11, 2026

Amazon Aurora PostgreSQL now supports direct querying of Apache Iceberg and Parquet data in your data lake

Amazon has introduced a new capability for Amazon Aurora PostgreSQL that allows users to query operational data alongside data stored in data lakes formatted in Apache Iceberg and Apache Parquet. This feature enables direct access to Iceberg REST Catalog-compatible catalogs, eliminating the need for data duplication or relocation. The integration of DuckDB within Aurora PostgreSQL allows users to query live operational data, including uncommitted writes, alongside data from data lakes using standard PostgreSQL syntax. This capability is available on Aurora PostgreSQL versions 17 (starting with 17.11) and 18 (starting with 18.6). Users must create an Aurora PostgreSQL cluster, attach an IAM role with the AuroraAnalytics feature, and enable the aurora_analytics extension to utilize this feature. They can create foreign tables referencing Iceberg or Parquet data in the data lake, and query them using familiar PostgreSQL commands. Aurora optimizes query performance through techniques like predicate pushdown and column pruning, and users can monitor query behavior with metrics provided by aurora_analytics_stat_statements(). Direct querying of Apache Iceberg and Parquet data is available across all commercial AWS Regions and AWS GovCloud (US) Regions without additional charges, aside from incremental Aurora compute and Amazon S3 request costs.
Designing a Concurrent Donation Ledger With FastAPI and PostgreSQL
Tech Optimizer
October 10, 2026

Designing a Concurrent Donation Ledger With FastAPI and PostgreSQL

Every year, billions of dollars flow through online crowdfunding platforms, leading to donor concerns about the allocation of funds. Traditional crowdfunding applications lack transparency, often leaving funds in unverified balances without real-time auditability or guarantees against double-allocations. GoodCause, a social impact platform, aims to address this issue by creating a system that ensures financial integrity and atomic fund distribution. Its architecture is based on mobile-first delivery, atomic auditability, and strict data isolation, utilizing technologies such as React Native, FastAPI, and PostgreSQL. The platform incorporates a mobile client for cross-platform navigation, a backend API for domain-driven routing, a database with Row-Level Security, and a payment gateway for secure transactions. A significant engineering challenge is preventing race conditions and double-spending in concurrent donation scenarios.
Supabase Announces $150M in New Funding and Turso Acquisition
Tech Optimizer
October 10, 2026

Supabase Announces $150M in New Funding and Turso Acquisition

Supabase has secured 0 million in funding led by GIC, with participation from Alphabet's CapitalG, IronArc, and SquarePeg. This follows a million Series F round four months earlier. Supabase has acquired Turso, a platform for managing databases for agentic workloads, to enhance its product development and provide liquidity for employees. The company is adding over 1 million users and 4 million databases monthly, with 70% created by AI-driven tools, reflecting a 600% year-over-year increase in databases. Turso's architecture allows for on-demand provisioning of databases in public cloud and BYOC setups. Glauber Costa, founder of Turso, will become Head of Agentic Services at Supabase. Supabase, established in 2020, supports over 13 million developers with a comprehensive backend solution.
How and why we built Managed PostgreSQL for AI workloads
Tech Optimizer
October 10, 2026

How and why we built Managed PostgreSQL for AI workloads

The Nebius Managed PostgreSQL team developed a solution for AI workloads using CloudNativePG and Kubernetes, resulting in improved backup and restore times by transitioning from Barman to WAL-G, reducing a 1.5 TB backup duration from over a day to two hours. Since its launch in 2025, the service supports hundreds of production clusters and manages tens of tebibytes of data. Key features include high availability, point-in-time recovery, automated failover, compressed and encrypted backups, rapid restoration, connection pooling, and strict synchronous replication for zero data loss. Kubernetes and CloudNativePG were chosen for their advantages, including out-of-the-box features and open-source distribution. Each cloud project has a separate MK8s cluster for customer isolation and security. Persistent data is stored on fault-tolerant SSD drives, enabling swift instance migration during failures. Regular physical backups are archived, allowing restoration within a seven-day window with a single click. The transition from Barman to WAL-G addressed slow backup and restoration times and resource control limitations. With WAL-G, backups and restorations for a 1.5 TB cluster improved to two hours and one hour, respectively. PgBouncer was implemented to manage database connections effectively, reducing connection storms during peak AI workload traffic. Two vector search extensions, pgvector and pgvectorscale, are available to cater to different memory and dataset needs. Benchmarking showed that Nebius performs competitively with AWS at a lower cost. The infrastructure integrates cloud engineering principles with specialized expertise to meet AI workload demands.
How Telangana’s MeeSeva migration to open source could make governments rethink software costs
Tech Optimizer
October 10, 2026

How Telangana’s MeeSeva migration to open source could make governments rethink software costs

In April 2026, Telangana's MeeSeva platform migrated from the Oracle database to the open-source PostgreSQL database after 15 years of reliance on Oracle. This transition aimed to reduce annual costs of approximately ₹10 crore associated with Oracle licensing and maintenance. The initiative began in 2023, and the migration was executed with less than four hours of downtime, processing around 80,000 transactions daily. The migration involved over 85 microservices, three Oracle servers, and 1,939 tables, with the process automated by the Ora2Pg tool, which handled 60-70% of the transition. The migration cost was below ₹10 lakh, and after the switch, MeeSeva processed nearly 200 million transactions without data loss. The new architecture includes automatic failover capabilities to minimize disruption.
Microsoft reveals plan to make Windows 11 Search much less painful to use, including a performance boost
Winsage
October 10, 2026

Microsoft reveals plan to make Windows 11 Search much less painful to use, including a performance boost

Microsoft is transforming Windows 11 Search to improve user experience by enhancing performance, reducing memory consumption, and streamlining the interface. The redesign is currently being rolled out to Windows Insiders for feedback before a broader launch. Key changes include a new architecture using WinUI 3, which improves performance and memory efficiency, and a simplified interface that merges results into a cohesive list. The updated Search will better understand user intent, support typos and synonyms, and include features such as inline previews and expanded file information. Users will also be able to execute common Windows tasks directly through the Search interface, such as switching to dark mode or turning on Bluetooth.
Microsoft’s AI-Powered Windows Strategy: 5 Things To Know
Winsage
October 9, 2026

Microsoft’s AI-Powered Windows Strategy: 5 Things To Know

Microsoft has transitioned its Execution Containers capability into general availability, creating a secure environment for AI agents on Windows 11. Nvidia's CEO praised Microsoft's MXC architecture as a revolutionary foundation for application development. Organizations are increasingly relying on managed service providers to navigate AI tools. The general availability of Microsoft Execution Containers allows organizations to define access for AI agents, with real-time policy enforcement. The open-source MXC library integrates across Windows, macOS, and Linux, helping manage agent interactions and mitigate risks. Microsoft is enhancing the Copilot experience with hybrid intelligence features and revamping Windows Search for better performance. Preorders for new Copilot+ PCs powered by Nvidia's RTX Spark technology are open, with shipments starting on October 16. Device partners like Dell, HP, and Lenovo are also launching AI PCs on the same date. Later this year, Dell and HP will introduce Nvidia’s DGX Station technology to enable local running of frontier-class AI models on Windows devices.
Search