* Posts by talk_is_cheap

68 publicly visible posts • joined 13 Jul 2011


Top cloud players reject Microsoft's attempt to settle EU licensing complaint


Re: The only chance the Euro companies have

This has nothing to do with the location of the server on which data is held, but rather the fees charged by Microsoft to install their software on the server and what other software you have to license at the same time.

MS, AWS, and Google all have EU-based services that conform to the data sovereignty rules you hint at. But it is often far cheaper to license Microsoft Server and Microsoft SQL when installing it on a node at Microsoft Azure than any other cloud provider regardless of location. Such pricing can be considered anti-competitive practices and so such cases as this can result.

Italy bans ChatGPT for 'unlawful collection of personal data'


Re: "there may not be a lawful basis"

That is how many European countries operate their legal systems. In the UK law is used to restrict what someone can do - so we have a law stating that we can not own submachine guns. Other countries grant the right to do something such as the right to own a submachine gun. So in the UK we have the right anything unless told not to, while elsewhere you can do nothing unless granted the right to do so.

Workday sued over its AI job screening tool, candidate claims discrimination


Re: Little more than "automated pseudoscience"

You are missing the key step when someone gives you a pile of 50+ CVs and tells you to pick 2 for interviews. You first remove at least half of them by only selecting the lucky ones - divide the pile into 2 without reading any of the CVs and put one pile in the bin. You are now left with the lucky ones.

Save $7 million on cloud by spending $600k on servers, says 37Signals' David Heinemeier Hansson


Other hosting options

There are a number of providers that do a good job of renting older physical servers. My 3 node VMWare development environment at ScaleWay with a basic firewall and S3 services has a month cost that is about the same as my single production EC2 instance on AWS.

It's not just the instance costs - many of the internet-based CI providers have high running costs as all the CI tasks run on AWS instances, such costs can be greatly reduced if they allow you to define instances on your own hardware.

Toshiba COO dumped over entertainment expenses scandal


China's Guizhou Province

To provide some context, China's Guizhou Province has a population of fewer than 40m people so that is one server per 10 people. The reality is that the report is just totally wrong 4m servers with a budget of $2.9b would mean no more than $725 being spent on each server and that is excluding all the rest of the infrastructure.

VMware turns 25 today: Is it a mature professional or headed back to Mom's house?


its been a long time.

I started to use it back in 98/99 in a development environment where there was a need for 1001 different test systems and VMWARE's tools on the desktop basically turned a rack space problem into a disk space problem.

I've been sticking VMware into sites ever since with the only problem time being when they decided that they were going to try and charge based on memory usage.

Even today I still use VMware, but the VMs are likely to be hosting whole systems defined as docker containers or k8s environments.

It works and it provides a common environment to all logical systems so I no longer have to worry about every new network card driver release for every deployment system and OS version.

China unveils massive blockchain cluster running homebrew tech


So they have built an audited database

Centralized databases with audited insertion are nothing new, with the most basic just having a field with a CRC value generated from the current and previous record. This allows checks to make sure a record has not been changed.

Blockchain tech is far more about doing all of this in a decentralized way, when you can't just throw together the fasted hardware that money can buy, on the fastest possible network, with the lowest possible latency.

I'm guessing that all the CPU power is just there to handle the cryptography calculations needed for the claimed peak 240 million tps, so the scale of the system may just be due to limitations in the CPU core design being used.

Prepare to be shocked: Employees hate this One Weird Clause


I once worked for an IT company where the standard contract has a re-written version of the UK copyright law that basically assigned all past or current personal rights I may have to the company. Oddly I never signed and after being around for about 4 years the company got in a new HR officer whose past experience was running staffing for hotels. A few weeks after joining she walked into my team's office and in front of my team informed me that I had to sign to contract by the end of the day or I would be let go. My team look at me and then look at her and just all started laughing, which got her a little confused, it got worse when I just told her that I would start clearing my desk. She never could get her head around the fact that IT staff were not as easy to replace as people who cleaned hotel rooms. She did not last very long and the next HR officer asked me to help rewrite the company contract so that it did not try and misrepresent basic British law.

Google dumps 12,000 employees after project probe


Re: Someone is conspicuously absent

They have not suffered from growth for many years and so are unlikely to have overhired - just overpaid for businesses that they have purchased.

IBM Annual Revenue

(Millions of US $)

2021 $57,350

2020 $55,179

2019 $57,714

2018 $79,591

2017 $79,139

2016 $79,919

2015 $81,741

2014 $92,793

2013 $98,367

2012 $102,874

2011 $106,916

2010 $99,870

2009 $95,758


On the 12th day of the Rackspace email disaster, it did not give to me …


Re: who's bollocks: Rackspace, or hosted exchange?

Office 365 is a completely different beast. This was built from the ground up to be a multi-tenant cloud service. I think it's very probable that Windows Server does not form any part of it, nor does Exchange. There will be data sanitization and security tripwires in every Internet-facing component, and between every internal component, and 24x7 security ops monitoring every aspect of the platform.

You would think that, but there are a lot of tasks that involve a remote powershell session to the Office 365 servers to allow you to make changes to Exchange settings that have not been exposed via the web GUI.

GitHub's Copilot flies into its first open source copyright lawsuit


Re: FOSS conditions

"shouldn't be too difficult to add something like"

Copyright does not work that way in most major countries - you do not have to detail that you retain copyright as you automatically own copyright over your own work. Instead, you grant people access to your work or if you are a developer working at a company you will have likely transferer ownership of your work to the company as part of your contract.

Backblaze thinks SSDs are more reliable than hard drives


For an operation, the size of Backblaze a failed boot drive would mean installing a new drive and just running cloud-init to re-install the node.

From the fact that they are using entry-level drives like the Crucial drive listed, it is clear that they have no concerns about the drives.

OVH to hike prices, blames 'l'inflation'


Re: OVH based in a major nuclear country has trouble with Russian natural gas prices ?

France has a number of reactors offline due to repair issues and many of the runs running are unable to run at full power because of a lack of water - they use local rivers for their cooling water and the lack of rainfall has impacted the rivers' flow rates.

After eleven-year wait, Atlassian customers promised custom domains in 2023


It is easy to beat 11 years - try a ticket started 1/Nov/2006

The issue is regarding the ability to change the author on a wiki server


Yes, that is right, also an Atlassian product issue - when someone writes a wiki page they are expected to remain with the company for life and beyond as there is currently no way to re-assign who is the 'author' of a page. The result is that any company that has used Confluence for any length of time can never track down the person who is currently maintaining a page.

Not much longer before this issue will be old enough to drink and vote in the UK.

Why Intel killed its Optane memory business


Intel trying to use it as a market 'advantage' did not help.

By making it an Intel only solution the only customers who could even consider it was those that were certain that their long-term plans would be based around Intel CPUs - any uncertainty on your future CPU needs and Optane based memory layers within your system design became a non-starter.

Vendors are hiking prices up to 30 percent and claiming 'it's inflation'


JetBrains sent out such a notification today.

It was very nice message along the lines of we have not put up prices for years, so we now plan to make up for it with a 30% rise in a few months time.

Former AMD chip architect says it was wrong to can Arm project


But why bother?

If AMD can advance their x86 based core design at the same speed as the ARM core design, there is little value in splitting their R&D budget and paying third-party additional IP fees. We already know that AMD is planning x86 cores with different levels of functionality and the only thing that seems unclear is how (or if) AMD plans to mix core types on a single CPU.

At last, Atlassian sees an end to its outage ... in two weeks


Considering the age of Jira and 16 year old tickets that are outstanding for complicated tasks such as changing the owner of a wiki entry I would say that NoSQL is right but in terms of pre-dating SQL rather than post-dating it.

UK arm of Sungard Availability Services goes into administration


The real issue is that their business model is currently dead in the water.

They offer

Colocation Servers - in a world where companies are moving more and more to cloud providers - why own your own kit.

Managed Hosting Services - again just part of the offering from cloud providers.

Workplace Recovery - WFH now offers much the same service.

What makes it worse is that any client that really needs these types of services are the ones that also need the latest spec environments - the power density of the latest top hardware normally exceeded what even a 3-year-old data centre can provide. Over the years I've gone from an average of 1A (230w) per U (of space) for my deployments to 2A, with 3A and 4A becoming more the norm. Such jumps mean new builds or lots of redoing power and cooling in old builds while leaving a lot of space unused.

International police shut down 15 server infrastructures as part of VPNLab.net's takedown


I guess that means that the phone networks will be next as they are used by scammers to perform serious criminal acts.

Pension cold-calling financial services biz cops largest ever fine from UK data watchdog


Re: £750 per referral, about £1 per person fine

The £750 would be the amount that EB would be paying the third parties if they provided a lead that resulted in a signup and a commission payment from whichever pension provider they got involved.

As for the conversion rate of cold calling - think 0.001% or less, hence the reason for outsourcing it to people who are paid such a high figure if they get a result.

The real cost is going to be felt by the pension providers if anyone was transferred to them without the correct advice having been given and followed.

Nobody cares about DAB radio – so let's force it onto smart speakers, suggests UK govt review


First, we need some content worth listening too and then we need a transmission standard that can deliver the content across the UK, rather than just in limited areas.

Many years ago Magic FM (also known as Tragic FM) was reported as having a playlist of only around 100 tracks which for many stations now would be considered a large track library. Magic/Tragic did not do themselves any favours as they published a Christmas CD box set with many of the tracks included - it allowed the iPod generation to rip the tracks and skip the adverts and chat - something that more and more people have done.

Today my phone has my whole CD collection on it with no compression loss. With 1,800 odd tracks of what I wish to listen to about the only time I use a DABs based radio is as my alarm clock as it also provides the news when it goes off.

Microsoft to charge $200 for 32 GPU cores, sliver of CPU clockspeed, 6GB RAM, 512GB SSD... and a Blu-Ray player


The GDDR6 used on the 2 systems may also have a speed difference, so a difference cost per GB and the Blu-Ray BOM cost is likely to be somewhat higher than $20. The real cost saving may come from the fee paid to AMD for the processor as the greatly reduced CU count will allow the use of a lot of dies that would have failed the testing process for the full-size system.

The real problem may come in the future if to many of the S series units are sold compaired to the X series. What will MS do if developers start to just deliver games that play correctly across both the S and X. The value of the X design will be devalued and the PS5 will become the platform for 'advanced' development.

Got a few spare terabytes of storage sitting around unused? Tardigrade can turn that into crypto-bucks


The court order goes to Tardigrade, not the person storing the data block(s). The end storage node is just holding a large number of encrypted data blocks with no way to unencrypt them. Even Tardigrade can not unencrypt the data, but they can remove them from the system and provide information regarding the creator of the blocks.

FYI: When Virgin Media said it leaked 'limited contact info', it meant p0rno filter requests, IP addresses, IMEIs as well as names, addresses and more



having spend some time trying to purchase a service from VM it is clear they could not run a bath let alone a piss up in a brewery. They make talktalk look like a well run business.

Firefox now defaults to DNS-over-HTTPS for US netizens and some are dischuffed about this


One question is who is more trustworthy?

My local ISPs (PlusNet and Virgin) who are under the control of the UK government or an ISP in the USA which is not directly under the control of the UK. For now I'll take the USA provider, with the added advantage that my requests are encrypted over the wire. I'm sure the UK government will at some point roll out a DNS platform that we must all use by law until then I am happy with having options.

Apple drops a bomb on long-life HTTPS certificates: Safari to snub new security certs valid for more than 13 months


Re: It's optional

The CAs are more worried about losing business from this change - If you have to keep updating your certs you are more likely to look at an automated solution and if you start to research that you end up moving to letsencrypt.org. The choice of CAs in most large companies is historc - remember all the marketing from Versign in the past and it was aimed at business people not techies. This change will allow the techies to put forward alternatives.

Atlassian anticipates $1.6bn total revenue for fiscal 2020 as subscriptions make it rain


Why would you expect a UK registered, US quoted company with US$862.3m of debt to be paying tax anywhere in the world, let alone Australian where the rate is set at 30%?

Leave your admin interface's TLS cert and private key in your router firmware in 2020? Just Netgear things


Just move to Origin Broadband

It is going to depend on what devices this hole is found on. The one they document seems to be a high end Smart WiFi Router. As the hole would allow the spoofing of the router you could grab the admin details. With these you could change setting on the router - so wifi access rights, firewall setting and VPN settings could open up wider security issues within any environment. Beyond that you have all the NETGEAR tools that turn the router into a file store and backup solution.

So to answer your question - with these certs what can't they do to a device and the network it is controlling and defending.

SoftIron's strategy to bring Ceph storage to the masses: 'A really, really sh*tty computer'...


"For starters, it reverses the traditional storage server layout by moving CPUs to the front of the PCB and storage drives to the back. This means cool air from the fans blows over the drives first, and then the CPUs – which wouldn't make any sense in a compute server."

Such a configuration does not make any sense as data centres are often configured as cold and hot aisles where the front of the equipment (air intake) is installed facing the cold aisle and the rear (air exhaust) facing the hot aisle.

From the few photos on Softiron's website, this seems to be a misunderstanding by the reporter as there is a photo showing drive storage behind the front panel and the motherboard at the back of the system.

When 2FA means sweet FA privacy: Facebook admits it slurps mobe numbers for more than just profile security


Re: Google too

A Three 123 SIM allows for a £5 topup on their website. I now have such a SIM as my second SIM in my phone. You just have to remember to use it every 6 months.

More nodding dogs green-light terrible UK.gov pr0n age verification plans


Re: Just like buying a magazine.

VPN software is being written to get past the great firewall of China. What chance do you think you have of blocking anything with a basic router.

Up the stack with you: Microsoft's Denali project flashes skinny SSD controllers


I guess the Linux world will have fun with this.

By the time that the zfs team and one day an improved Btrfs team start to develop to arrays of such SSDs we may/will see some nice advantages from moving parts of the stack to the OS and main memory. As for Microsoft, all I would expect is a lot of work to tie systems to their OS while they would not make the investment in their file systems to take full advantage.

Let's Encrypt updates certificate automation, adds splats


Re: "...admins will have to edit a DNS record to prove..."

>> How many DNS hosting providers have mandatory multi-factor authentication on their web portals?

Well dnsmadeeasy offers it as an option.

A print button? Mmkay. Let's explore WHY you need me to add that


Just move to Origin Broadband

Yep, it is quite often the Dinosaurs in a company that want such features. The reason why they are still at the company is because they are often the owners or board members. So if the software is to be purchased, deployed and retained it better come with a print button.

Veeam's reverse-IBM, the rebrands, the new hires, and... DRAM, what's that, Samsung?


A single 8GB HBM2 package consists of eight 8GB HBM2 dies, which are vertically interconnected using over 5,000 TSVs (Through Silicon Vias) per die.

I guess that should be eight 8Gbit HBM2 dies, otherwise it would be a 64GB HBM2 package.

Want a new HDMI cable? No? Bad luck. You'll need one for HDMI 2.1


With the Audio Quest Diamond HDMI Cable - 5.0 Meter down to just £3,000 on Amazon I wonder what price they will try charging for the latest standard?

China plots new Great Leap Forward: to IPv6


Re: Well, you can see the attraction of it!


Chinese state security (translated): "We enforce an unchangeable IPV6 address (by law) on all devices used inside the Great Firewall and bingo, we can follow everyone, everywhere and log everything they say, do, read and connect to. No hiding behind NAT any more. What's not to like?


No because devices will still move between locations, each of which will be a different IP network.

BT hikes prices for third time in 18 months


Just move to Origin Broadband

Their basic phone plus adsl2+ service is currently only around £12 a month if you sign up for a year. It works and if there is an issue they have a 0800 UK based support team, who do seem to have a clue about what they sell.

Making money is so DRAM easy for some memory-flingers


The players who are left in this market can just claim that any expansion plans will take far too long to deal with cost issues during this market cycle - so extending the cycle. Unless one of these announces a major investment none of the other have to bother. It's hard to prove the operation of a cartel if all the producers just sit on their hands because no one has made a commitment to expanding their production.

Wikibon drops bomb, says Intel's Optane could be a flop...tane


We are still waiting for the 'real' product

Currently, all Intel seems to have to show is a 32GByte stick that is aimed at caching. Something of a dead end product considering the performance caches of many U.2. sticks are now larger.

The original pitch for this tech was 256GByte memory modules for installation into large fat servers where each Intel CPU could address 8+ memory sticks. The possible resulting I/O was meant to interest the big data and VM markets.

Today I wonder how many designers are more focussed on what they can do with 2 AMD top end processors and 128 PCI 3 lanes - that's a lot of 2TB U.2. sticks without paying premium rates to Intel for a lot less storage capacity.

PCIe speed to double by 2019 to 128GB/s


AMD interconnects

It starts to become very clear why AMD is happy to just use PCI links as its interconnects for its new server chips. Currently, the design uses 64 PCI 3.0 links to connect 2 processors which is as far as their published designs go. With PCI 5.0 they could greatly increase the overall interconnect speed or allow a 4 processor design using 16 links* between each with the same overall performance as now. I guess PCI 4.0 will provide a shorter term compromise with 4 possible processors connected at a higher overall speed, but lower processor to processor speeds.

* A 4 processor system would only need 48 PCI links (16 to each of the other 3 processors), this would allow the possiblity of hyper-cube designs (many processors not directly connected to each other) if AMD designs the correct protocols for cross processor communications.

HPE Labs manufactures monster memory Machine system


I think this is more a show case for the "X1 photonics module"

On paper, this is an advanced inter-link solution that can shift 150GB data per second in a local area configuration.

Reg now behind invisible HTML5 Bitcoin paywall


The real joke is that it's not a bad idea if people are allowed to op-in to such a system. It would also need to be coded to use GPU cycles rather than CPU cycles if it's to generate any meaningful results.

I need an ISP that offers IPv6. Virgin Media: Whatevs, nerd


Re: Non Story

Note true, they have removed the option for new customers to have 5 static IP addresses with their business service (for an extra £5 a month).

All in all their service while fast has never be configured to offer a high level of service to business users.

Microsoft: We're hiking UK cloud prices 22%. Stop whining – it's the Brexit


Re: £

And Microsoft lowered the cost of all its software back in November 2015 when the Pound hit €1.42 as during 2012 its was only worth around €1.20

Brit ISP TalkTalk scraps line rental charges


The real kick in the teeth is the contract term.

They seem to think that they can get current long term customers to tie themselves into an 18 month contract. Before we could get a discount on the line rental for a 12 month contract and often if you talked about leaving a discount on the broadband service as well.

M.2 SSD drive format is under-rated. So why no enterprise arrays?


So why no enterprise arrays

One rather key issue is that you are talking about the 4x PCI standard, with the latest devices being able to fully load the 4 PCI lanes in test conditions. Such performance is able to overload 10Gbit ethernet links and saturate a 40Gbit link. This in many ways just makes them too fast for use in large arrays. Why bother if you can just deploy current SAS based options.

5 of them placed on a 16x PCI could make a nice RAID 5 solution, but will need a very high performance control chip to be developed, with performance specs well beyond anything currently available. For every 1GB/s of write performance the controlling chip will have to also read 1GB/s from the array, xor it and then write the resulting parity information as well as the original data. While such a configuration may provide very high speeds the enterprise market seems to be more focussed on DIMM based SSD solutions which link directly to the server's CPUs. These remove the controller chip issue and will distribute the storage across all the CPUs in a system if required.

The other great use for these types of drive would be as caches within arrays built with slower storage devices, but currently the write endurance is not designed for such tasks. The 960 PRO 2TB only has a 1.2PB endurance, which is not much in a enterprise write cache deployment.

Quantum is shutting down sync'n'share biz Symform in July


I used this sevice for a year or so, but

Even before Quantum purchased Symform the writing seemed to be on the wall as it had a wide range of limitations that did not seem to be getting addressed. Once Quantum was in charge it got even worse as nothing happened.

I moved across to hibiC, which does the job of syncing files to the internet and between systems at very low cost. So all in all its good enough.

Intel, Warner lock horns with hardware biz over HDCP crypto-busters


Where's the rest of this pirating solution they're claiming everyone uses.

Due with the new graphics cards from AMD and Nvidia as we will all need some way of quickly re-encoding an uncompressed 4K60 data stream that has been captured by some yet to be released capture card that can handle a ~2GByte per second hdmi 2.0 data stream. There will also be the need to store the data stream to disk, so I guess rather a large and fast SSD sub-system will be required.