Modern flash storage relies on NAND memory cells that trap electrical charge, with newer 3D NAND designs stacking cells vertically and packing multiple bits per cell to boost capacity while lowering cost per gigabyte. These improvements have made flash ideal for SSDs and portable devices, since it lacks the moving parts of hard drives, giving lower latency and better durability. However, flash remains fundamentally different from RAM, serving as persistent storage rather than the fast temporary memory a system needs while running.
Google has issued new guidelines requiring Android app developers to reduce memory consumption, giving them until February 2027 to comply. The move responds to a global memory shortage that has pushed phone makers to cut RAM in entry-level devices, often dropping from 6-8GB to just 4GB. Apps exceeding Google's memory thresholds for dynamic RAM use, bitmap handling, and DEX code will be throttled or force-closed by the OS.
Politico sources report the Trump administration is drafting another round of tariffs on semiconductors, aiming to push chipmakers toward US-based production. Commerce Secretary Howard Lutnick reportedly favors letting companies import chips duty-free up to a threshold linked to their domestic investment, though the exact terms remain unsettled. The move would layer new costs onto PCs, consoles and data centers already strained by memory shortages.
Android developers are grappling with growing memory constraints as more apps integrate on-device AI features, which require substantially more RAM than traditional app functions. This shift is pushing hardware and software teams to rethink memory management strategies to keep devices responsive.
Apple's newly unveiled Mac mini with M6 and M5 Pro chips carries several specs the company didn't spotlight at launch. Memory bandwidth actually varies by RAM configuration, topping out only with the M5 Pro model, and the machine includes an Apple-designed N1 chip enabling Thread smart-home support. Additionally, units will ship with the unreleased macOS Golden Gate already installed.
SK Hynix held a groundbreaking ceremony at Purdue University in West Lafayette, Indiana, for its first U.S. facility, a $4 billion plant dedicated to packaging high-bandwidth memory chips. CEO Kwak Noh-Jung said the site will become a key American production base for HBM by 2030, with its first cleanroom expected to open in October 2028, though actual chip manufacturing will still occur in South Korea and China.
Cloudflare engineers made five successive changes to how DNS cache entries are stored in memory for Big Pineapple, the platform underlying 1.1.1.1 and related DNS services that hold over 250 billion cache entries at once. The changes shrank the per-entry memory footprint by more than half, freeing roughly 100 terabytes across the company's servers while also boosting insert throughput by 43% and cutting lookup latency by 19%.
Google published a memo detailing enforcement of new per-app memory limits introduced with Android 17, first rolling out on Pixel devices. Apps that exceed the thresholds will be slowed down or terminated, and starting in February 2027, non-compliant apps could face reduced visibility and publishing restrictions on the Play Store.
Google announced two new Android app quality requirements in response to memory chip shortages fueled by AI data center demand. One mandates stricter thresholds on dynamic memory and bitmap usage along with code optimization to prevent crashes and slowdowns, while the other requires apps to automatically restore sign-in status when users switch devices, via the Zero Tap Sign-In standard. Developers have until February 2027 for the memory rules and April 2027 for sign-in compliance.
Google announced that starting February 2027, Android apps and games on the Play Store must comply with new per-app memory limits based on a device's total RAM, with games given more lenient thresholds than apps. Titles that fail to meet these caps risk reduced visibility in the Play Store or being blocked from publication altogether. The rules build on per-app memory limits first tested on Pixel phones under Android 17, now expanding device-wide.
Samsung unveiled a 16GB LPDDR5X memory package that runs computation directly inside the chip, achieving 614 GB/s of internal bandwidth between its DRAM banks and compute units. That figure is eight times higher than the 76.8 GB/s the same package can push through its external pins, according to a Tom's Hardware report from the conference.
A developer has released a lightweight database designed specifically for AI agent memory, built on a typed schema with supersession rules instead of relying on semantic text retrieval. The system aims to run cold-first at the edge, avoiding the cost and complexity of maintaining hot vector indexes in RAM or SSDs for smaller, resource-constrained deployments. The project was shared as a Show HN post on Hacker News.
GoKawiil