Monthly Archives: September 2019

Linux 5.4-rc1 Kernel Steps Forward With Next-Gen GPU Bits, Arm Laptop Support & exFAT


It’s coming one day late due to the last minute entropy/RNG patches to improve the random behavior during boot time (among other late patches), but Linus Torvalds has just tagged Linux 5.4-rc1 as what will be the last major stable kernel release of 2019.

See our Linux 5.4 feature overview if missing it from the weekend. Linux 5.4 is bringing new GPU support from Intel and AMD Radeon, long-desired Microsoft exFAT file-system support, Icelake Thunderbolt support is finally in place, Qualcomm Snapdragon 855 SoC support, the ability to run mainline kernels on various Arm consumer laptops, better IO_uring support, Logitech Lightspeed receiver support, and a heck of a lot more. See our feature list for all the details.

Linux 5.4 stable is likely to debut by the end of November or potentially December should last minute bugs come about. Following that will be the Linux 5.5 cycle before the holidays but that will not be out as stable until into Q1’2020. Linux 5.5 in turn is what will hopefully be shipping in the likes of Ubuntu 20.04 LTS.

Linux 5.4-rc1 is available from Git and while awaiting any release announcement from Torvalds. At least as of -RC1, the Linux 5.4 kernel is still carrying over the codename of the “Bobtail Squid”, a relative to the Cuttlefish. Linux 5.4 benchmarks coming!

New LPWAN Applications Expand Use Cases | IT Infrastructure Advice, Discussion, Community

Long viewed by enterprises as a solution for connecting far-flung fixed devices, use of Low-Power WANs (LPWAN) is expanding into asset tracking of mobile items from pets to parcels.

In addition to helping companies better manage their supply chains and provide visibility into core processes, researchers claim health monitoring, and smart(er) agriculture are also emerging as fertile ground for low-speed IoT applications.

LPWAN is an umbrella term that covers a variety of established technologies that can be used to support long haul IoT applications comprised of devices such as sensors. They’re connected to inexpensive radio units that use batteries to transmit small amounts of crucial data to a central location and can last ten years.

See also: Why Low-Power WANs Are Emerging as a Strong Option for IoT Apps

Tracking solutions are targeted at automakers, freight shipping firms, and logistics service providers. With asset tracking, enterprises pay per attached tracking device and in some cases, the amount of data tracked. They can get a dashboard and can often add data analytics and integrations with core company business applications.

Covering your assets

When it comes to shipping and transportation, supply chain managers are looking to asset tracking to prevent loss, rapidly address delays that could become costly, optimize routes for faster deliveries, and address the risk of theft and degradation of assets that require special environmental conditions be maintained. Many of those lacking necessary visibility are small and medium businesses (SMBs).

As is the case with virtually all asset tracking IoT apps, the goals are to track the location of the asset and monitor its contents (as is the case with shipping businesses). It’s argued that current asset tracking approaches for supply chains can be costly and cumbersome, making the LPWAN approach attractive to those seeking visibility without breaking the bank.

Using an LPWAN along with low-cost rental devices is aimed at keeping costs down and providing supply chain visibility to have-nots. In shipping, for example, a sensor is often attached to every pallet, container and/or package in a truck. Buying them in large volume can result in a sizable capital expenditure.

NB-IoT asset tracking

In launching its narrowband-IoT-based (NB-IoT-based) asset tracking solution with RoamBee in July, T-Mobile estimated 77% of firms have limited to zero visibility into their supply chains. The solution is a Platform-as-a-Service offering featuring a combination of licensed capacity on the carrier’s LPWAN, $10 a month per-device charge and includes data.

NB-IoT LPWANs – which are also offered by AT&T and Verizon – should prove a solid match for asset tracking as they support a maximum data rate of roughly 200 kbps, which is greater than competitors pitching networks using Sigfox (100 bps) and LoRa (50 Kbps) technologies.

With its NB-IoT solution, T-Mobile includes unlimited transmissions at up to 64 kbps upload and 20 kbps download.

Advantage GPS?

Though typically more expensive than NB-IoT, Global Positing System (GPS) provides benefit not available from network operator-provided LPWAN services and solutions.

Wireless operators claim to provide nationwide NB-IoT networks. However, there are areas where they have limited or no coverage. Satellite systems-based GPS offers blanket coverage. NB-IoT networks, like T-Mobile’s, don’t support roaming, so coverage remains an issue. Finally, GPS tracking is available beyond the U.S., which is a big plus for shippers whose freight crosses country boundaries.

Together, these features are why many transportation and shipping companies have long relied on GPS-based asset tracking, which is more robust than what’s offered via LPWANs.

The human asset

In healthcare, talk of patient monitoring using LPWAN services is gaining interest after forward-thinkers decided that smartwatches should be the device through which health is monitored.

Smartwatches will be a strong entry point as an estimated 17% of U.S. broadband households own a smartwatch, according to researcher Parks Associates. “The low-power aspect of LPWAN will help extend the battery life on smartwatches, a key demand among smartwatch owners, while still maintaining strong connectivity,” according to its recent report.

Making agriculture smarter

Though not an IoT application that requires a mobility element, connecting crops and livestock in agriculture is seen as fertile ground for LPWAN use if it can be offered in a way that farmers with limited technology use and big expenses can cost-justify.

In agriculture, it’s envisioned that LPWANs would be used to collect data on the health of the animals, their location, and when they give birth. Purported benefits are improved yields, a higher-quality product, and greater insight for farmers to more efficiently manage their operations – including optimizing irrigation efforts.

If offered in an affordable manner, which means a low cost of entry for small and midsize farms, LPWANs should prove alluring for asset tracking. By 2024, over two million farms and 36 million cattle will be connected in IoT, according to ABI Research, a market advisory firm in a recent report.

Related Network Computing articles:

Taking AI to the IoT Edge

Talking IoT Standards with the Open Connectivity Foundation


Source link

Linux 5.4 Should Improve NUMA Hugepage Allocation Performance


It turns out Linux 5.3 shipped with potentially subpar performance for the allocation of hugepages but that should be rectified in the now open Linux 5.4 cycle for trying to provide a sane default allocation strategy on NUMA boxes.

With Linux 5.4, the kernel will now avoid the reclaim operation when compaction isn’t likely to succeed. It will also allow hugepages to fallback to remote nodes when seeking hugepages via madvise but the node-local hugepage allocation failed. Additionally, Linux 5.4 reverts two patches from Linux 5.3 that ended up regressing other workloads unintentionally.

Ultimately, if all goes well and long story short, Linux 5.4 should be offering a more sane default allocation strategy for hugepages on NUMA systems. That in turn should help performance. More details via this Git merge that happened on Saturday.

Five Considerations for a Wi-Fi 6 Deployment Plan | IT Infrastructure Advice, Discussion, Community

The saying: “Proper preparation, prevents poor performance,” holds true for technology deployments.

Whether you’re ready for an upgrade to the next version of Wi-Fi in your enterprise or venue or not, a plan for Wi-Fi 6 (aka 802.11ax), is only as good as the effort IT managers put into it. There’s no race to upgrade, despite numerous benefits, but there are considerations for those prepping for a deployment.

The greatest benefit of Wi-Fi 6 is that it handles more traffic more efficiently, meaning it can shoulder a heavier load than its widely deployed predecessors. Better still, the new Wi-Fi standard is backward compatible with earlier versions, letting users avoid stranded investments and flash cutovers.

Products emerging

Vendors continue to push products down the Wi-Fi 6 product pipeline. They include access points (APs) and routers, as well as clients. The vendor ranks include Aerohive, Aruba, Asus, Broadcom, Cisco, D-Link, Intel, Netgear, Qualcomm, and Samsung.

Some verticals will embrace the new standard earlier than others. But don’t expect a stampede once certified Wi-Fi 6 product become available late this year or early in 2020. That said, enterprises can plan now for a deployment, using the five key considerations below.

Are you congested? Though it offers large increases in data speed, Wi-Fi-6 is all about efficiency in wireless networks. With the number of attached devices soaring, the new standard was designed to deal with network congestion.

A congestion-combatting technology is uplink and downlink orthogonal frequency division multiple access (OFDMA), which increases efficiency and lowers latency for high demand environments. Another core feature is called multi-user multiple input, multiple output (MU-MIMO). It allows more data to be transferred at one time, enabling APs to handle a larger number of devices simultaneously.

Device support. While more Wi-Fi 6 APs and routers are coming through the product pipeline, devices such as smartphones, tablets, laptops, and more need to progress for implementors and end-users to enjoy the purported efficiency, lower power consumption, and speed benefits of Wi-Fi 6. Also, on this list are IoT devices, which are key to smart home efforts, and units for broader, outdoor applications.

Devices supporting Wi-Fi 5 or earlier standards will still work with the newer infrastructure, however.

Certification timeline. It looks like potential early implementors might want to wait a while, as the certification testing that the Wi-Fi Alliance will perform has been pushed back from its original target of the third quarter of this year (now), until yearend. It’s not clear how long the certification process will take. As a result, it’s looking as if certified products won’t be available until 2020.

Non-certified products can be an option but consider, for example, that WPA3 security is required for Wi-Fi 6 certification, but there’s a chance it wouldn’t be included in unapproved equipment.

The business of technology. What’s the business case for a Wi-Fi 6 deployment? It’s often difficult to justify spending on new IT infrastructure without a solid ROI. But that’s where consumer-first applications can continue to drive wireless technology implementations.

Enter the media, sports, and hospitality verticals. Wi-Fi 6 is already envisioned as delivering higher-speed streaming content services to consumers, at a price and likely with ads and sponsorships. That’s a clear ROI.

Many sports venues have been challenged to keep up with fan demand for more and faster wireless to access and share video clips and more on social media. But forward-thinking venue owners have used data analytics to learn more about and better engage with their customers. The new version’s efficiency and congestion-addressing features should help meet increasing demands for speed, data, and performance.

Wi-Fi in the hospitality industry has evolved from a nice-to-have perk to a must-have one, to one that must be challenge-free, or the customer experience goes from dependent to dissatisfied. It can be argued that providing a superior customer experience is one step removed from revenue generation. It can also be argued that a dissatisfied customer is a departing customer. Quality of experience is core to the hospitality industry.

Design considerations. It’s recommended that in preparing for Wi-Fi 6, firms continue the practice of designing high-quality 20 MHz channels. Planners must also remember that the new standard specifies lower power consumption for the devices connecting to networks with improved efficiency, especially sleep/wake time optimization. This benefits the mobile phones, laptops, or other IoT devices.

Also, remember to budget power as Wi-Fi 6 networks require even more power for the businesses and organizations deploying them. This makes power budgeting increasingly important as you add 802.11ax APs with more powerful processors and additional radio chains.

And when an upgrade plan progresses, don’t forget to save time for extensive field testing and validation as these functions are crucial to assembling high-quality and performance networks.

With efficiency-focused Wi-Fi 6, proper planning should lead to much more than just a smooth deployment.

Source link

It Turns Out CPU Speculative Execution Can Be Useful For Random Entropy / RNG


While CPU speculative execution has caused a lot of frustrations over the past two years due to the likes of the Spectre vulnerabilities, it turns out CPU speculative execution can be exploited to be a viable source of random entropy for random number generators.

Particularly on newer Intel/AMD CPU microarchitectures where speculative execution is much more advanced than hardware from years ago, it’s been found that measuring the execution time of loops relying upon speculation is random enough to be a cheap and speedy source of entropy. Straight-forward loops without any extra instructions obviously don’t work out but with adding “useless” instructions in the mix does yield non-deterministic execution times when measured via RDTSC (reading the CPU timestamp counter).

Longtime kernel developer Thomas Gleixner who has been involved with a lot of the Spectre/Meltdown mitigation work was the one presenting his preliminary code and research on the topic. It’s not necessarily as strong as some of the more rigid sources of entropy data, but it’s quick and seems to work well particularly with newer CPUs as outlined in this mailing list post.

Linus Torvalds commented and he believes that this is not very reliable and a simple jitter entropy implementation. But he did post his own proof-of-concept code for improving the jitter entropy code based upon this.

We’ll see where this work heads and if a future Linux kernel code could be making use of speculative execution for another source for random number generation.