Tech Field Day is now part of The Futurum Group. Our events bring together innovative IT product vendors and independent thought leaders to share information and opinions in a presentation and discussion format. Independent bloggers, freelance writers, and podcasters have a public presence that has immense influence on the ways that products and companies are perceived and by the general public.
"Cloud Control Plane... is a really difficult part of Cloud Computing" - Oxide has people that know all about it since one of the mentioned previous control planes once rebooted a whole datacenter (unintentionally, that is). That's a feature I don't think they'll be shipping again very soon.
If the S24 Ultra has a total of 2 antennas, wouldn't EMLSR mode achieve higher efficiency performance compared to STR mode? This is because EMLSR can utilize all NSS for transmission using the link with the same wireless communication.
What I'm curious about is the power consumption of clients in the real world, where they will generally support EMLSR. However, I observed in packet sniffing that the S24 Ultra has STR capability during the connection phase. I found this very interesting. How many antennas does it have for each independent radio? This STR feature can cause significant battery consumption. Unfortunately, detailed WiFi chip and hardware specifications for clients are not mentioned.
Hi, Does the STR-supported S24 Ultra phone have 2 radios and 2 antennas? When operating in STR Mode, are the antennas configured as 1x1 for the 2.4GHz band and 1x1 for the 5GHz band to communicate with the AP via MLO?
We built such solutions more then 10 years ago at a previous job I held and we just did not call them edge compute. It had central orchestrator where one could manage distributed firewalls, SDWAN, cameras, file sync, VMs, IoT, containers, and a host of other features. The nice thing about EVE-OS is that it is OSS and I hope it continues to build the critical mass for wider adoption. I would prefer though to see KVM in the architecture of EVE-OS instead of Xen for the VM orchestration. Just a personal preference.
Caching data in Hazelcast and managing its synchronization with a database is a common pattern to improve application performance. The data in Hazelcast (in-memory cache) provides fast access times, while the database provides persistent storage. The timing of when data is committed to the database depends on the caching strategy you choose. Here are some common strategies: Write-Through Caching Write-Behind Caching Read-Through Caching Cache-Aside (Lazy Loading)
Thanks for this! I really _HATE_ that people think on-prem is expensive, we have all our compute power on-prem and its _CHEAPER_ than having it "in-the-cloud"!
Good conversation - albeit quite depressing to see the same conversations that have existed in the industry for the last few decades being repeated. It was good to see paul try and get the conversation moving on security. Software will not be secure until the industry cops punitive regulation - just as what has happened to many other industries over the years - food, automobiles, aircraft etc. At the moment, when a breach which exploits software vulnerabilities occur, the customer feels the pain, the customers customers feel the pain and the software development companies are not even mentioned - at most, they might lose a couple of customers. There simply is not a strong enough business case for SW development groups to properly secure their software or supply chains and there won't be as long as we continue to allow the industry to hide behind software license contracts and the argument that to put more focus on security will somehow inhibit innovation. Just the fact that SW companies are having kittens about having to produce an SBOM says it all - An SBOM can be compared to the ingredients listed on a 50c can of beans, but apparently, customers asking to have an idea of what is actually in the software causes very real headaches.
You can watch the entire presentation at techfieldday.com/appearance/oxide-computer-company-presents-at-cloud-field-day-20/ and if you really want to make sure you didn't miss a second the live recording is on our LinkedIn page
A big issue is how long these projects take to get off the ground when using the "antiquated" infrastructure of today. That initial phase of hardware/infrastructure acquisition forces the business into a corner to keep said project, even if the project was a bad decision. After all the investment in the building/BOM phase, people are apprehensive about "failing fast" on a bad project (sunk cost fallacy).
Such a refreshing way of thinking. Dragging the benefits of the cloud to the on-prem customers and also using telemetry in the system to help the customer and support staff. This type of technology also helps keep the cloud honest because it gives customers a genuine option as at the moment it's Cloud or Bust.
