In this week’s issue of README Highlight (#42, 2023), we are taking a look at the following project: OpenTofu. OpenTofu is an open-source tool for building, changing, and versioning infrastructure safely and efficiently. It can manage existing and popular service providers as well as custom in-house solutions.

The key features of OpenTofu are:

• Infrastructure as Code: Infrastructure is described using a high-level configuration syntax, allowing it to be versioned and treated like any other code. It can also be shared and re-used.

• Execution Plans: OpenTofu generates an execution plan before making any changes, showing what it will do when called. This helps avoid surprises and allows for better control over infrastructure manipulation.

• Resource Graph: OpenTofu builds a graph of all resources and parallelizes the creation and modification of non-dependent resources. This ensures efficient infrastructure building and provides insight into dependencies.

• Change Automation: Complex changesets can be applied to infrastructure with minimal human interaction. The execution plan and resource graph help operators understand what changes will be made and in what order, reducing potential errors.

The OpenTofu repository contains the OpenTofu Core, which includes the command line interface and the main graph engine. To learn more about compiling OpenTofu and contributing suggested changes, refer to the contributing guide. Bug reports and enhancement requests can also be submitted following the same guide.

If you find a vulnerability or potential vulnerability in OpenTofu, please follow the Security Policy for reporting it. The project uses the Mozilla Public License v2.0.

For more information about OpenTofu, you can visit their Manifesto and their About page. You can also join their Slack community or check out their weekly status updates on the project’s GitHub repository.

The Mixtile Cluster Box is a server enclosure designed for small business applications and edge computing. It consists of four Mixtile Blade 3 Pico-ITX single board computers (SBCs), each powered by a Rockchip RK3588 processor. The SBCs are connected to the enclosure via a 4-lane PCIe Gen3 interface through a U.2 to PCIe/SATA breakout board.

The Cluster Box has been recently released by Mixtile, following the company’s work on the software and technical details. It is available for purchase on Mixtile’s website for $339, excluding the SBCs.

The specifications of the Mixtile Cluster Box include support for up to four Mixtile Blade 3 SBCs, each with up to 32GB LPDDR4 RAM and up to 256GB eMMC flash storage. The enclosure also features a control board running OpenWrt 22.03, with a MediaTek MT7620A MIPS processor, 256MB DDR2 system memory, and 16MB SPI flash storage.

The Cluster Box includes an ASMedia ASM2824 PCIe switch with four PCIe 3.0 4-lane ports. It also provides storage interfaces through four U.2 breakout boards, with four NVMe M.2 M-Key slots (PCIe 3.0 x2 each) and four SATA 3.0 ports. Networking capabilities are offered through a Gigabit Ethernet port.

The enclosure is equipped with two 60mm fans for cooling and a power button with a blue LED indicator. It is powered by a 19 to 19.5V/4.74A power supply through a DC jack. The dimensions of the Cluster Box are 213 x 190 x 129 mm, and it is made of a metal case with SGCC steel materials. It has an operating temperature range of 0°C to 80°C and a storage temperature range of -20°C to 85°C. The relative humidity ranges from 10% to 90% during operation and 5% to 95% during storage.

Users can access the Mixtile Cluster Box through OpenWrt using SSH or a web interface. The Rockchip RK3588 boards come preloaded with a customized Linux system with Kubernetes. Control of each Mixtile Blade can be done through OpenWrt using a command called “nodectl,” which allows users to list active nodes, rescan nodes, power on/off nodes, reboot nodes, flash firmware, and enter the console of a specific node.

For more technical details and a getting started guide, users can refer to the documentation website provided by Mixtile.

Overall, the Mixtile Cluster Box offers a compact and powerful solution for building a four-node server cluster with Rockchip RK3588 SBCs. With its PCIe connectivity, storage options, and OpenWrt software, it provides a versatile platform for various server, Linux, DevOps, and home lab applications.

Source: CNX Software – Embedded Systems News.

SD Association (SDA) has announced the new SD 9.1 specification that doubles the speed of microSD Express memory card speed up to 2GB/s, defines four new SD Express Speed Classes to ensure guaranteed minimum sequential performance levels, and adds support for multi-stream access and related power and thermal management to assure the guaranteed performance.

The new SD 9.1 specification doubles the speed of microSD Express cards, with a maximum speed of 1,969 MB/s achievable over a PCIe Gen4 x1 interface. The cards remain backward compatible with the existing microSD card standard at the lower UHS speeds. The specification also introduces four SD Express classes with minimum guaranteed speeds.

Three new features have been defined in the SD 9.1 specification: power management through maximum power values, thermal management with specific thresholds, and an access rule for multi-stream recording.

While several companies have announced microSD and SD Express cards, they are not yet commercially available. It remains to be seen whether these types of SD cards will be widely adopted.

The press release and white paper provide additional details about the new features added to the SD 9.1 specification.

Source: CNX Software – Embedded Systems News.

Prusa3D has recently released an online course for their popular 3D printer, the Original Prusa MK4. This course is aimed at 3D printing beginners who have chosen the MK4 as their first printer, as well as experienced users who want to refresh their knowledge.

The MK4 online course follows the same pattern as its older relatives, starting with unpacking the printer, running the setup wizard, and printing the first sample model. The course also explains the printer mechanics and provides information on connecting the printer to the internet via Prusa Connect. Additionally, the course covers the basic controls and menus of the PrusaSlicer app, introduces the Printables.com model database, and provides an overview of 3D modeling using Tinkercad and Fusion 360.

The course is presented in a unique hybrid text-based form, which combines the advantages of both text and video formats. The content is color-coded, split into small chunks, and includes plenty of pictures and short videos. Quizzes are also included to test the learner’s knowledge. Upon completion of the course, learners can generate and download a certificate.

The MK4 online course is available for free with the purchase of the assembled variant of the MK4 or for members of the Prusa Education program. Existing customers who have already bought the assembled MK4 will automatically receive the course. To access the course, users can log in to their Prusa Account and click on the “My Courses” option. For kit variant owners, the course can be purchased for a nominal price of $5 or with Prusameter bonus points.

In addition to the MK4 course, Prusa3D has also made major updates to their existing courses, bringing them up to date with the latest version of PrusaSlicer and Printables.com features. The courses are currently available in English, with translations expected to be completed by the end of October.

Prusa3D has also made improvements to their e-shop, allowing users to mix virtual and physical products in the same shopping cart. They have also added recommendations for minor products that can be purchased together with a printer. Furthermore, it will be possible to buy multiple copies of the same course at once, making the process more convenient.

In the future, Prusa3D plans to release a series of 3D print postprocessing courses and is considering a dedicated course for their Original Prusa XL printer.

AMD has announced the release of the Ryzen Threadripper 7000 series, bringing significant updates to their Threadripper processors. The new series offers up to 96 Zen 4 cores, requiring RDIMM memory for Threadripper platforms moving forward. This update caters to both high-end desktop (HEDT) enthusiasts and professionals.

Compared to the previous Threadripper PRO 5000 series processors, the Ryzen Threadripper PRO 7000 WX-Series is a substantial upgrade. While the previous series topped out at 64 cores and 128 threads, the new series allows for up to 96 cores and 192 threads, matching the core counts found in AMD’s EPYC Genoa processors. The Zen 4 Threadripper processors can clock up to 5.3GHz, offer up to 384MB of cache, and include features like AVX-512 and 128 PCIe 5.0 lanes.

The Threadripper PRO 7000 WX-Series parts can reach up to 350 Watts. The AMD Ryzen Threadripper 7000 series processors are in a league of their own at the top-end. Intel currently does not have any Core i9 Extreme Edition processors, and their much-delayed Intel Xeon W-3400 series only goes up to 56 cores. Even if one were to consider the Xeon Scalable “Sapphire Rapids,” there is still a core count limitation. The Threadripper 7000 series processors are expected to provide an interesting battle against the Xeon Scalable processors, especially the Xeon Max.

The AMD Ryzen Threadripper 7000 series is divided into separate platforms for PRO and HEDT. The Threadripper PRO platform, with the WRX90 chipset, offers PRO manageability features, 8-channel memory support, 148 PCIe lanes, and support for Threadripper PRO processors. In the HEDT space, there is the TRX50 chipset, which lacks PRO management features, supports only four-channel memory, has 92 PCIe lanes, and is compatible with both PRO and HEDT Threadripper processors.

The Ryzen Threadripper 7000 series (non-PRO) includes processors with up to 64 cores. The Threadripper 7980X, with a 350 Watt TDP and a maximum turbo boost of 5.1GHz, leads the non-PRO line-up. Other options include the 24-core Threadripper 7960 and the 32-core Threadripper 7970X.

Source: Phoronix.

SiFive has introduced two new IP blocks, the Intelligence X390 NPU and the Performance P870 RISC-V core, which are designed for SoCs targeting Generative AI and ML applications.

The Performance P870, previously covered by CNX Software, is a high-performance core that supports the RVA23 RISC-V profile specification, Vector 1.0, and Vector Crypto. It features a six-wide, out-of-order 64-bit core and can be configured with up to a 32-core cluster. The P870 offers over 12 SpecINT2k6/GHz and a 50% peak single-thread performance improvement compared to the previous generation SiFive Performance P670. It also includes SiFive features such as 2x 128b VLEN RVV, vector crypto and hypervisor extensions, IOMMU and AIA, non-inclusive L3 cache, and RISC-V WorldGuard security. The P870 is fully compatible with Google’s platform requirements for Android on RISC-V and supports Linux.

The SiFive Intelligence X390 NPU is built on the U7-Series core with a 64-bit RISC-V ISA and an 8-stage dual-issue in-order pipeline. It supports the RISC-V Vector extension and features a 1024-bit VLEN, 512-bit DLEN, single/dual vector ALU, and the Vector Coprocessor Interface eXtension (VCIX) for direct connectivity of external hardware accelerators. The X390 NPU offers a 4x improvement in vector computation compared to the Intelligence X280 NPU, thanks to its single-core configuration, doubled vector length, and dual vector ALUs.

SiFive envisions the P870 and X390 being paired in powerful RISC-V SoCs with up to 32 cores and 8 AI accelerator cores (along with optional custom accelerators) to enable generative AI in data centers. The company promises a full AI solution with optimized libraries, as well as OpenXLA and NN models. Additionally, the Performance P870 is suitable for DPU, storage, networking, 5G applications, and automotive SoCs. In the automotive market, the P870-A RISC-V core can be integrated with the Intelligence X280 AI accelerator and SiFive S7-A cores for ASIL B/D compliance, supporting applications such as standalone ADAS, central compute, and in-vehicle infotainment (IVI).

SiFive also suggests that the P870 and X390 may find their way into consumer applications like edge AI security cameras, mobile devices, wearables, and smart TVs, although no specific details were provided.

For more information, please visit the product page and press release.

Source: CNX Software – Embedded Systems News.