BlackBerry QNX’s Post

High Performance. Real-time Execution. Unrivaled Safety and Security. And all without compromise! Introducing the QNX Software Development Platform 8.0, powered by the next-generation QNX OS. Why QNX SDP 8.0 for next-gen systems? ✅ Gain high performance with seamless scalability and real-time execution. ✅ Boost development cycle efficiency for faster time-to-market. ✅ Innovate with state-of-the-art tooling and cloud-enabled versions of core QNX products. Curious about how QNX SDP 8.0 can transform your automotive development journey? Explore the possibilities at https://bit.ly/3Awg0SZ #BlackBerryQNX #EmbeddedSystems #SoftwareDefinedVehicles

Heard all the hype of HBM bundled with the CPU? While there are lots of reasons HBM could be beneficial, whats not beneficial is buying a platform, locked with expensive memory on a small per core basis. Imagine in the HPC space wasting expensive memory to support standard cluster operations, filesystem support etc. That performance comes at a cost. On top of that, in already tight deployment scenarios where the demands of computing are growing exponentially, the luxury of wasted rack space is a thing of the past. On the flip side, the new Genoa-X combines the amazing density of our Epyc platform, with a massive L3 cache. How does that “stack” up? You can get the performance boost of ultra highspeed cache, and combine that with standard DDR memory and maximum core density enabling you to have not only a blazing fast processor memory combination, but also a flexible platform as well. Workloads of a wide variety can deploy side by side. Amazing as it is, we are just getting started!

❓ DID YOU KNOW that the latest release of the QNX microkernel, QNX® SDP 8.0, has numerous improvements that make it particularly suited to HPC applications and HPC-based systems in key areas? These include: ✅ Next-generation microkernel tailormade for HPC ✅ New and updated kernel schedulers ✅ Hard real-time determinism ✅ Uniform parallelism and pre-emption ✅ High-performance networking These advancements position QNX SDP 8.0 as the premier choice for developers looking to fine-tune their HPC environments without compromising on safety and security. Don't just take our word for it—experience the power of QNX SDP 8.0. Request your free trial today: https://bit.ly/3Awg0SZ #HPC #BlackBerryQNX #Innovation #Microkernel #RealTimeSystems #HighPerformanceComputing #TechInnovation #SecureSystems

Time to load up on more AMD and get short INTC stock?

Some Context to the post from Hot Isle: Earlier this week, TensorWave published an article demonstrating compelling results comparing the MI300x with an H100. Intrigued by their findings, we endeavored to assess the performance of our #mi300x under similar conditions. With some minor adjustments and leveraging open-source tools, we achieved even slightly more improved results. We extend our gratitude to TensorWave for sparking this exploration (and presenting us with a challenge ;-))! Stay tuned as we continue to advance our Paiton tool, promising unprecedented outcomes for LLM inferencing (details here: https://lnkd.in/eKA6gq8D). Meanwhile, we are embarking on a similar analysis using open-source tools with the Llama 3 70B model :) Check out Nscale for more upcoming info!

The AV1 open standards-based codec is used to compress a video file or stream to a smaller file size while preserving and maximizing the visual quality. EIZO set out to build on the research published by NVIDIA to prove the use of AV1 for military applications. Dive into all the details of our testing completed with our Condor™ AGX-IOX product that utilizes the #NVIDIA Jetson AGX Orin™ to compare AV1 with H.264 and H.265 codecs. We sought to measure latency in addition to the bandwidth and quality metrics. Download the technical study “A Better Way to Encode and Decode: A Technical Study of AV1 Video Coding Performance”: https://hubs.li/Q02wGFF30 #AV1 #Video #EmbeddedComputing #OpenStandards #VideoProcessing

Webinar: Understanding MPSoC Real-Time Processing Americas Wednesday, August 16, 2023 Time: 10-11am (PDT) 11am-12pm (MDT) 12-1pm (CDT) 1-2pm (EDT) Registration link: https://lnkd.in/gkdYev7b In this webinar: We will explore the Arm® powered real-time processing capability of the UltraScale+ MPSoC and the hardware and software requirements for implementing time sensitive applications. Topics include: Capabilities of the Arm Cortex®-R5 System level implementation decisions on Zynq UltraScale+ devices Arm Cortex-R5 programmers model, exception handling, memory model and memory management. #amd #xilinx #mpsoc #soc #fpga #webinar #arm #cortex #r5 #zynq #ultrascale+ #realtimeprocessing #timesensitive #memorymanagement

Kontron introduces the VX6096