Eugene Kovko’s Post

eXtremeDB’s xFlash implements a copy-on-write (CoW), page-mapped recovery mechanism to ensure database consistency and durability delivering reliable, lower-latency transactions directly on NAND flash without the overhead of complex file systems. Learn more: https://t.ly/sAZXH #extremedb #databasemanagement #embeddedsystems #nandflash

Critical Section (CS):A section of code that must be executed by only one thread at a time to prevent data inconsistency. Race Condition:Program correctness depends on thread scheduling → can cause unpredictable results. ✅ Solution: Locks or atomic operations. Preemption:When the scheduler interrupts a running thread to give CPU to another. Controlled by locks, disabling interrupts, and priorities.

❌ An SSD that doesn’t show up can bring any workflow to a standstill. 7 Data Recovery Experts created a YouTube tutorial that shows: ✔️ Why SSDs sometimes disappear ✔️ Step-by-step fixes that really work ✔️ How to protect your data during recovery ▶️ Full guide: https://lnkd.in/djy5w9fa

New release Ultralytics v8.3.202 | TFLite INT8 per-channel quantization fix 🚀 More accurate YOLO11 TFLite INT8 models with per-channel quantization and leaner artifacts; plus faster distributed tuning and steadier CI for smoother edge deployments. Minor updates: ✅ Per‑channel INT8 fix and disabled `batchmatmul_unfold` reduce TFLite artifact size for edge devices ✅ Distributed Tuner seeds immediately when MongoDB collection exists, improving multi‑worker starts ✅ CI stability: macOS runner pinned to macOS‑26 for consistent builds Ultralytics v8.3.202 release notes ➡️ Release v8.3.202 https://lnkd.in/dmgGD7xs

This is a re-post with new picture. I had modified the RISC-V picture without changing the bottom portion to reflect x86. ϕEngine has only 4.5× the code density of x86. The 19× was for RISC-V. Sorry about that. We compared RISC-V with ϕEngine on saturated add. But how does the x86 compare? At least it has flags, so the code won't be quite as bad as RISC-V. For an even comparison, we will assume that the calling convention allows passing of the arguments in EAX and EBX: 0000: 01 d8         add  eax,ebx 0002: 71 07         jno  b <NoSat> 0004: 72 06         jb  c <Minus> 0006: B8 FF FF FF 7F     mov  eax,0x7FFFFFFF 000B: c3           ret 000C: B8 00 00 00 80     mov  eax,0x80000000 0011: C3           ret 0012 Keep in mind that, when overflow happens, the sign is the opposite of what it should be. The code came to 18 bytes. That is 4.22× times the code density of RISC-V. Much of that is because it has flags and can detect overflow in hardware. But the x86 can't do clamping in hardware so it has to test the flags and load the appropriate values into the register that returns the result. So ϕEngine code density is 4.5× that of the x86. The x86 is about midway between RISC-V and ϕEngine on a log scale.

How Leo singleton storage will work under the hood (we're still working on it): storage count: u32; ↓ Compiles to: mapping count: bool => u32; Access: let x = count → Mapping::get(count, false) Fixed key (false) simulates singleton behavior.

A clarification on the behavior of blosc2.where(): it returns a lazy array, deferring the actual computation. In contrast, np.where() computes immediately when a Blosc2 array is passed as an argument. Thanks to Olivier Grisel for pointing this out. This notebook may help to clarify the situation. This also highlights how compression can improve performance.

StorageReview.com recently tested our expanded hardware lineup and confirmed what sets Object First apart: powerful data resilience delivered with simplicity. Read the full review: https://bit.ly/3HguDzX

The Transactional Flash Translation Layer, or TFTL™, is what enables eXtremeDB/rt to deliver reliable ACID-compliant database transactions directly on unmanaged flash memory devices, without the use of a latency inducing file system.  Learn more: https://t.ly/dOjz7 #flashstorage #databasedesign #littlefs #embeddedsystems #RealTimeData

Simple LZW Compression Algorithm Implementation in C A straightforward C code of the Lempel-Ziv-Welch (LZW) compression algorithm in C. LZW is a popular lossless data compression method that substitutes repeated data occurrences with references to a single instance. #C #lzw #algorithm #compression