This is lowkey one of the most interesting videos they have done on this channel in a long time, in terms of tangible, world-changing technology. The utility of that AI search tool is so unfathomably insane
Google's Gemini let's you search through all of youtube. You can just ask it for random things and it will find youtube clips of it. Truly a game changer.
@@0BlueauraOh yeah because that's not problematic at all, the AI definitely only "catches criminals" and doesn't harvest your entire life's data, it's definitely not going to create profiles of completely innocent people that could be abused in the blink of an eye. No way there could be severe infringement of privacy rights while the presumption of innocence that prevents this is completely being undermined. Surely we wouldn't get false positives all day long because the data the AI is trained with is full of human bias, most prevalent: racism. There's zero chance that such technology would create a system of total control that could be abused to erect an authoritarian regime in an instant. All of the above definitely wouldn't happen and it's not like have ample precedence from all around the world where those fears already became true. You literally prepare society to be easily taken over by a fascist dictator with this. You make sure that people are completely defenseless.
As a senior data center engineer in a multi billion dollar enterprise I really enjoy such content (with a bit of envy), while we still have to deal with the old school netapp metro clusters, HPE XP8 or IBMs SAN volume controllers you guys can play with the innovative stuff, even if its totally overkill for your usecase.
I get it now. You need two of everything for proper redundancy. Elijah can drop stuff when Linus isn't there. Elijah can mess stuff up when Dennis isn't there. Good hire, LMG.
As someone who built software like Weka for ~8 years, I appreciate how excited you are about these insanely complex systems and the performance they bring.
This is easily one of the most beneficial and practical uses of AI I've seen. Eliminating the tedious work of sifting through thousands of hours of content to find something relevant? Yes, please.
My home desktop cannot lose data as well - in the sense that if something were to fail, I'm too broke to afford redundant storage. Enterprise reliability is nuts, it's great to watch :)
you'd be surprised at how cheap redundancy can be. i got a 16 tb external i use to backup anything important/excessively large for like 150 dollars. you can pick up a cheap 1tb external for like 30 dollars or less on sale, and that's pretty much all you need to back up your really important shit unless you do video editing/raw photo editing.
@@Bierkameelthose systems can be implemented with small numbers of nodes also, they are not by definition better than this, but in large enterprises are built at scale and spread across racks or even separate buildings. You could do that with this setup too, they just don't need to.
The best part is - he probably didn't know a lot about it, but he figured it out. Also, the storage isn't really redundant since it's in the same rack, in the same building, in the same continent :D It never ends!
There are so many terms or protocols involved when actually dig into 'networking' , they can't really delve too far into them since are (mostly) very niche. Stuff like going with SMB should be explained more in a video like this.
@@benwu7980 if I really want to know more about something I can just give it a goog. These videos are perfect the way they are right now. I think they’re really fun. I’m a simple man; if Jake is excited in a video then I’m excited watching it.
I must say, Linus Tech Tips has truly outdone themselves this year! The improvement in content quality is simply outstanding. It's incredibly noticeable how much work you've put into enhancing everything from writing to video production. The videos now are on a whole new level of coolness! Guys, you were good before, but now you're next-next-next level! Kudos to the LTT team for continuously raising the bar and delivering top-notch content.
Okay… the server was neat but HOLY COW that sorta AI search program thing at the end was unbelievable. One of the coolest products I’ve ever seen. That’s an actual game changer for folks who have a lotttttt of data and through which it’s not easy to search.
That Epyc CPU could hold the entire system memory of a Win 2000 machine in its L3 cache. Can you even imagine how fast it would be if your entire kernel AND workload is in CPU cache, and main memory isn't even being used!? That's wild.
I’m so glad you finally made this video. I have always wondered how the editors find relevant archive footage. You’re telling me it was just from memory this whole time?!
"What could go wrong?" "I mean, a lot." Ah, it's always a good sign when those words are being said on the _other_ side of the screen 😂 Glad to see Jake back again! Had been thinking it's been a while
THE MAGIC OF BUYING TWO OF THEM: like all proper nerds of this vintage, the idea of anything being dual is just ruddy exciting: dual cpu, dual hard disk, dual gpu, dual dual server with dual dual dual dual psu…
@@invisi1407 I think he's the one that originally began calling it "the _magic_ of buying two of them"? Unless he's quoting it from an even older show...
The one thing I really take away from this one: "New-new-new-new-new-new-new-new-new-Whonnock goes REAL BRRRRRRRRRRRRRRRRR!" The amount of knowledge and expertise involved in these projects exceeds my level of computing soooo far at this point^^ Still fun to watch!
Great to see professional server stuff shown to the public. One small correction though: even though the ConnectX-6 cards have 2x200G ports, the card itself is limited to a total of 200G by its chip, and both ports have to share the total bandwidth.
I remember the first video from Linus Tech Tips that got me hooked to the channel - the first Petabyte project. Even after all these years it is still as entertaining as it ever was to watch these guys do what they enjoy. Thanks to Linus Media Group, I am now pursuing a career in IT. I feel grateful to have come across this channel because it has formed who I am as a person now. Keep up the good work.
To Add to the split brain talk, you need an odd number of nodes to guarantee they can reach quorum. Running only 3 nodes also creates issues when there is a failure as you now can't reach quorum and need to immediately get the node back up. When running 5//7/9 nodes outages are not nearly as urgent .
My guess is when you have more servers, the load on each is decreased, such that occasionally one can take a backseat. So with an even number of servers, by this logic, the failure of one server would not compromise the process and it would function as normal since the “extra” would not be needed right away and thus have more time to get resolved and back up and running. Networking allows the servers to handoff tasks to neighboring servers in the network. This is also why it was mentioned in this video that “two entire servers” can go down and not cause an issue and nobody would even notice. Also, the odd number discrepancy to check for biases, that’s usually an ideal for if you have multiple tasks computing at once that put the network of servers at full load, but otherwise an even number will still work fine as one is backup.
I just decommissioned something similar at work about 2 months ago these are awesome and super reliable. The ones we decommissioned were used for high availability to radiology records in a major hospital, ran for 7 years with no issue and was only replaced for a standard life cycle upgrade of the system!
@squidwardo7074 in our case yes. Because of the high availability of data it would take in our case 6 simultaneously failing servers to cause an outage. Or 2 simultaneously failing access switches with 2 simultaneously failing links for each server connected to each switch
Last time you lost data it was not because of lack of failsafes, but rather the fact that no one was keeping an eye on things. This may prevent failures when something goes wrong, but only if you actually respond to those failures in time. Make sure to have some sort of alert system this time around.
They have actual sysadmins now fyi, like not just some of the editors or whatever happen to also be good at this stuff (such as emily for example) but like they've been hiring actual dedicated IT staff
@@mysteryboyeeThat may be, but monitoring is something that they should probably cover at some point in a video like this. Redundancy is not just about hardware. And having sysadmins does not change anything besides the fact that they would know to setup ways to monitor the situation, unless they sit on a chair on front of the server rack at all times.
@@danielberglv259One would presume that dedicated sysadmin staff would indeed have it as part of their job description to actually do such things. They would be unfathomably stupid not to.
I REALLY want a Floatplane extra for this which dives DEEP into this stuff. I don't care if it's 3,5 hours long like the FP Exclusive of Kyle. THIS. IS. AWESOME!
I absolutely love these server oriented videos! Please never shy away from doing more! I almost wish you guys had a specialized channel for this kind of stuff and just let Jake geek out. Great stuff, LTT!
...the ending is pure gold: "..see all the lights, they never stopped blinking." :D such a powerful statement, so true for any debugging process no matter who you are!
@@vffaa few months ago/last year supposedly their TH-cam metrics said they were losing viewer attention during the intro, so they stopped adding it in. It was mentioned on the wan show. However personally I feel it gives the LTT a character like a proper TV show. I love it!
this is a great video to show how AI could be useful in an SMB - Small to Medium Business, not the shared drive :D I kinda wish you hadn't buried it 20 minutes into the video, but you had to, so we could see the technology that makes it reasonable to do. This is actually one of the most relevant videos of yours for me in what I do in my professional life. Good work.
15:43 Jake is referring to the fact that the latency accounts for the fact that they are using DPDK (Data Plane Development Kitty) to facilitate RDMA (Remote Direct Memory Access): basically running RDMA through software, and accounting for the latency as a result (I.e, in the userspace rather than at a kennel level). This is pretty fantastic!
That time jump at the beginning hooked me like a fish and I watched the whole video I love infrastructure content and other business related IT! I could watch it all day.
20:40 OK that is definitely the best feature so far, this era of information there are lots, but to actually get one that is relevant is the true power.
I'd love a video on the evolution of Whonnock over time. I think that's good for learning how to evolve our home server storage over time. Fun fact: one of the Whonnock videos was the first LTT video I ever saw.
It would be kinda sick to see an educational series on IT engineering and deployment given all of the amazing professionals working on projects like this on the front and back end.
The concept of cluster voting was developped by Digital Equipment Corp in the 1980s for its VMS clusters. The votes and quorum mechanism did not handle just failure of a node, but more importantly prevented a node that left the cluster from continuing to function (thus proventing it from accessing drives without enforcing locks/synchrinization with the other nodes). (called cluster partitioning). In the case of a 2 node cluster, you could designate a drive as quorum disk and give it votes, so any node in the cluster that still saw that/those drives would get the votes and could then continue. (in VMS clusters, drives could either be served by a node (and accessed from other nodes via network) or be standalone with direct hardware links to each node. (Digital developped extention to SCSI (called DSSI) to allow multiple hosts to access the same SCSI bus so multiple nodes coul write directly to the same physical drives). (later on, I beleive SCSI did get that functionality). The freezing of a node to prevent partitioning is very clitical to prevent a node that lost its connection to the other nodes from continuing to write to disks or perform tasks without the synchronization with the other cluster nodes. (in a cluster, there was a single lock database so when a process on a node too a lock on a file (or area of file) or a resource, that lock would exist on all nodes so processes on other nodes would not get that lock until first one released it). From a failover POV: You can have 2 nodes doing work since the locking is synchronized, so when one node fails, the other one continues and new network connectiosn all go to the second node instead of being split between the two. The other way is to have node 1 take and get lock and do all the work. Node 2 requests the lock but is put on hold because lock is unavailable. When node 1 goes down, node 2 gets the lcok and then processes all the work. Tandem NonStop (used for mission critical stuff such as Interac/credit card) has different fault tolerance. In the same chassis, a process would run on a single CPU/RAM, but a copy would run on a designated CPU/RAM, but basically have it writes to disk/network disabled. Both processes got the same data from network/disk. So the backup was a copy of the main one, and shoudl main one go down, the backup tok over right where the backup failed since the backup had identical RAM, process state and conections to network etc. However, while VMS clusters could be spread across multiple buildings (and up to 96 nodes), the Tandem Nonstop was within a single chassis in 1 computer room. so fault tolerant within chassis, but not disaster tolerant. A mere smb file server just blindly executes writes/reads from any node to any area of disk/file and when you have different windows instances accessing the same time, the last one writing to it wins.
"They don't know what HA means" Me.. working in IT for 25 years: "HA means nothing if its not distributed and works as a failover cluster in a DR event" TLDR; It's still not secure, it's not protected against floods, fires, or other disasters. It still lives in a closet. The system needs to be distributed physically before it can be called HA.
@@nickfarley2268 That's not true, the team could just be moved to a temporary location with remote access to the files... that's the whole point of a DR plan. In the event his whole team dies during the flood, sure... you're right, but that would be quite a flood :) A strong power surge could take down his whole "high availability" setup... not exactly HA.
Excellent comment. Really excellent. The serious stuff for HA is not interesting for a general audience. Is pretty boring, even for us IT professionals. Is not "cost vs risk", it doesn't have the features for being redundant and secure. Is like building "the world's most powerfull PC" with $200, and then argue that well... it has prioritized costs... A recomendation? Just do it right, and if it is not right for a video, don't make a video about it, period.
@@maxhennessy6676 sure, but that doesn't make it HA... that's my point. You can't say "it's a discount HA", since there is a real name for it... just a failover cluster. Even if it load balances, still not HA. HA has many requirements and physical distribution is one...
Jake loses weight with every video we see him in. If this is intentional, keep up the good work! If it isn't, I pray you stabilize toward a healthy goal weight. ❤
Now, for true redundancy, do a nightly mirror/backup over fiber optic or microwave link to an identical setup in the labs building. Call it Whonnock 10_2. God forbid another UPS or power bus catch fire in the server room. Or Linus pull 3 servers for a fail over demo by mistake.
Using Ceph you can define your failure domain and run a cross datacenter cluster. The failure domain is scalable, so right now I run a disk level failure domain at home, but in a bit I'll move to host level failure domain (once I add a few more hosts)
Not good enough..l they need backup in another part of Canada or US. That area will someday have a big earthquake off the coast and that will sever fiber optic lines and cause major power outages, etc…
19:43 first time ever hearing about Genomics Lab storage needs :). The to go example is always video editing / visual FX, because it's more familiar of course !
Seeing the software made me really wish for someone to build a consumer storage classifier. Like just that part, you could point it to a file in your regular file system, give it somewhere to store all the generated information, and be able to manage all your pictures and videos.
My mind was blown by the storage/infrastructure itself... then they started talking about that INSANE search ability, and I can't even. It's SO FREAKIN' COOL.
One thing I love about these server videos is that, unlike the high end PC builds you make, I might actually interact with one of these! Obviously not of my own, but as a professional software engineer deploying stuff for an employer. Even just making decisions based on what hardware bottlenecks will appear for different types of services! And the best part is not having to spend your own money on the outcome 😂
yeah, given that they use TrueNAS most of they time so probably won't be going with super expensive Weka long term, or similar competition. He did mention TrueNAS cluster but TrueNas says it is based on Gluster and that support is degraded at this point. I fell Ceph would be most appropriate, and I REALLY would like to see the video for how they pull that.
I am a small homelabber, and I love only these types of videos from LMG....I watch all of their server room/ servers/ storage kind of videos. I first thought they were gonna use Ceph but looks like Weka is super cool too.
Next video: "we lost all our data"
😂☠️
Server dies, new server installed. All other servers replicate from the new one instead of the other way round. Poof.
Wouldn’t be the first time 😂
Totally agree! 😅
😭
This is lowkey one of the most interesting videos they have done on this channel in a long time, in terms of tangible, world-changing technology. The utility of that AI search tool is so unfathomably insane
yah
Seriously this is the stuff I saty on the channel for
Google's Gemini let's you search through all of youtube. You can just ask it for random things and it will find youtube clips of it. Truly a game changer.
@@0Blueaurano thank you please
@@0BlueauraOh yeah because that's not problematic at all, the AI definitely only "catches criminals" and doesn't harvest your entire life's data, it's definitely not going to create profiles of completely innocent people that could be abused in the blink of an eye. No way there could be severe infringement of privacy rights while the presumption of innocence that prevents this is completely being undermined. Surely we wouldn't get false positives all day long because the data the AI is trained with is full of human bias, most prevalent: racism. There's zero chance that such technology would create a system of total control that could be abused to erect an authoritarian regime in an instant.
All of the above definitely wouldn't happen and it's not like have ample precedence from all around the world where those fears already became true.
You literally prepare society to be easily taken over by a fascist dictator with this. You make sure that people are completely defenseless.
“How’s the edit going? I’m holding your file server” was the best line 😂. And then when Jake walked over with a second node….*chefs kiss*
the answer was better "is it on wifi?" XD
it is still warm
@@39zack yeah, they'd be close😂
Gives the feeling of "hey, how's your blood doing? I'm holding your heart right here"
@@39zack is it on wifi is so freaking funny to come up with for how fast linus just ambushed him on camera lol
"I'm holding your server"
"Is it on Wi-Fi?"
Absolutely amazing
Let's be honest, that is quite a reasonanble thing to ask in that situation, I would do the same
bless u mark
"this, Jen, is the internet"
"Oh. Why are there no wires!??"
"It's wireless!!"
@@MocSomething Gotta remember to view that episode every once in a while. Thanks.
More importantly, I am totally not grounded with a rack and I want to hug it with my ears. ZAP!
As a senior data center engineer in a multi billion dollar enterprise I really enjoy such content (with a bit of envy), while we still have to deal with the old school netapp metro clusters, HPE XP8 or IBMs SAN volume controllers you guys can play with the innovative stuff, even if its totally overkill for your usecase.
THE INTRO HAS RETURNED. Thank you Dan.
It's not done until it's dan
I cry happy tears a little every time
I missed it
Isn't it in every longer video?
that and Linus' beard, it truly feels like 2018 again
I get it now. You need two of everything for proper redundancy.
Elijah can drop stuff when Linus isn't there.
Elijah can mess stuff up when Dennis isn't there.
Good hire, LMG.
You need to hire third to get full HA set !
cannot lose data, until Linus somehow manages to drop it eventually
14:05 Linus' face when everything is gone
Beat me to it lol
i mean they're ssds, a drop should do basically nothing
...
right?
if you know about databases, you can absolutely drop data!
I mean hopefully if he drops it only the 1s will fall out since they weigh more than the 0s.
As someone who built software like Weka for ~8 years, I appreciate how excited you are about these insanely complex systems and the performance they bring.
This is easily one of the most beneficial and practical uses of AI I've seen. Eliminating the tedious work of sifting through thousands of hours of content to find something relevant? Yes, please.
My home desktop cannot lose data as well - in the sense that if something were to fail, I'm too broke to afford redundant storage.
Enterprise reliability is nuts, it's great to watch :)
This not enterprise but small business junk.
Enterprise storage is dedicated like Netapp, EMC Unity, 3par and many others.
@@BierkameelTell me more
you'd be surprised at how cheap redundancy can be. i got a 16 tb external i use to backup anything important/excessively large for like 150 dollars. you can pick up a cheap 1tb external for like 30 dollars or less on sale, and that's pretty much all you need to back up your really important shit unless you do video editing/raw photo editing.
100GB of Google is $2/mth
@@Bierkameelthose systems can be implemented with small numbers of nodes also, they are not by definition better than this, but in large enterprises are built at scale and spread across racks or even separate buildings.
You could do that with this setup too, they just don't need to.
Videos like this are proof that Jake is a very valuable employee. Guy knows his IT stuff.
And he gets to play with the newest stuff like toys ;)
Yep! Yvonne's husband's boyfriend is pretty great at his job!
And god bless him for trying to make every server in that company Linus proof.
@@ChristopherHallett LOL 😂
The best part is - he probably didn't know a lot about it, but he figured it out.
Also, the storage isn't really redundant since it's in the same rack, in the same building, in the same continent :D It never ends!
My favorite LTT videos are the ones with Jake explaining networking that I do not understand at all
There are so many terms or protocols involved when actually dig into 'networking' , they can't really delve too far into them since are (mostly) very niche. Stuff like going with SMB should be explained more in a video like this.
Absolutely agree because that is also me
Rly not that difficult
@@kreuner11 sick dude
@@benwu7980 if I really want to know more about something I can just give it a goog. These videos are perfect the way they are right now. I think they’re really fun. I’m a simple man; if Jake is excited in a video then I’m excited watching it.
I must say, Linus Tech Tips has truly outdone themselves this year! The improvement in content quality is simply outstanding. It's incredibly noticeable how much work you've put into enhancing everything from writing to video production. The videos now are on a whole new level of coolness! Guys, you were good before, but now you're next-next-next level! Kudos to the LTT team for continuously raising the bar and delivering top-notch content.
He has a full team that coordinates this for him; it does really pay off to have a great team to make content together.
See you guys in a year when the server loses data
Okay… the server was neat but HOLY COW that sorta AI search program thing at the end was unbelievable. One of the coolest products I’ve ever seen. That’s an actual game changer for folks who have a lotttttt of data and through which it’s not easy to search.
Object recognition using machine learning isn't exactly a new concept, but I guess no one did an implementation like this before? 🤷♂
I wish TH-cam had a search like that. Some times I'm searching for something to use as joke or whatnot and only "popular content" appears.
@@nunoaguiar2525 a search feature like that implemented into youtube would be huge, people would cry over ai being used, but it would be huge
@@nunoaguiar2525It will take a long time, but I'm sure we'll reach that point in the future 5-20 years with youtube search.
@@wasituzayer9728object recognition was cool, but scene recognition is off the charts
That Epyc CPU could hold the entire system memory of a Win 2000 machine in its L3 cache. Can you even imagine how fast it would be if your entire kernel AND workload is in CPU cache, and main memory isn't even being used!? That's wild.
Yes, BUT.
It still needs to hold the necessary data and operations as well, otherwise you will run around with your shoe laces tied together.
I think it would hold the entire system memory of a modern FreeBSD or Debian machine, certainly it would for Alpine.
Windows 95 used something like 16MB, and could run on less IIRC
If you get the stacked cache versions you can get just over 1.1GiB of L3 cache per CPU. The new AMD stuff is nuts.
@@defeqel6537 Minimum was 4MB, recommended mimimum was 8MB.
I’m so glad you finally made this video. I have always wondered how the editors find relevant archive footage. You’re telling me it was just from memory this whole time?!
Lol pretty much.... For most of our history we've been a very small company. A dedicated person to ingest and apply metadata was not realistic. - LS
Probably mostly your memory Linus...@@LinusTechTips
@@LinusTechTipsthat sounds like a hard thing to do
"What could go wrong?"
"I mean, a lot."
Ah, it's always a good sign when those words are being said on the _other_ side of the screen 😂
Glad to see Jake back again! Had been thinking it's been a while
Jake looks fantastic and I'm so proud of him and the effort he has clearly put in. Good on you man.
THE MAGIC OF BUYING TWO OF THEM:
like all proper nerds of this vintage, the idea of anything being dual is just ruddy exciting: dual cpu, dual hard disk, dual gpu, dual dual server with dual dual dual dual psu…
I can't read that first line without thinking about Alec from Technology Connections. 🤣He's said that so many times in his videos. :D
@@invisi1407 I think he's the one that originally began calling it "the _magic_ of buying two of them"? Unless he's quoting it from an even older show...
Technology Connections gonna sue LTT at this point
watc interet anarchist
@@WackoMcGooseI think that was just all Alec. :D If you google the sentence, the whole first page is basically references to T.C. :D
8:25 linus gesturing with the server sent chills down my spine.
I thought only I was freaking out
My arm instinctively twitched as if to reach out and catch it
@@chuckpoe5297 this is what you get when money and goods mean nothing to him...he simply doesn't give a dam..
fr
u can even see his arm shaking
Dad, and son nerding out over servers. Warms my heart
Having an engineer like Jake around can make a world of difference for your company
The one thing I really take away from this one: "New-new-new-new-new-new-new-new-new-Whonnock goes REAL BRRRRRRRRRRRRRRRRR!"
The amount of knowledge and expertise involved in these projects exceeds my level of computing soooo far at this point^^ Still fun to watch!
You're killing me linus 8:27 like holy you made my heart drop.
Stop kissing his ass.. hi Linus 😂
@@UnExile huh?
It makes me so happy seeing the intro in so many new videos. It's the best intro on youtube, and screw viewer retention, I will watch it!
My words brother. It should be herecall the time it is legendary
Great to see professional server stuff shown to the public. One small correction though: even though the ConnectX-6 cards have 2x200G ports, the card itself is limited to a total of 200G by its chip, and both ports have to share the total bandwidth.
They did mention the total bandwidth limitation on those cards.
I remember the first video from Linus Tech Tips that got me hooked to the channel - the first Petabyte project. Even after all these years it is still as entertaining as it ever was to watch these guys do what they enjoy. Thanks to Linus Media Group, I am now pursuing a career in IT. I feel grateful to have come across this channel because it has formed who I am as a person now. Keep up the good work.
Was this a sponsored comment?
These networking server infrastructure videos are my guilty pleasure
To Add to the split brain talk, you need an odd number of nodes to guarantee they can reach quorum. Running only 3 nodes also creates issues when there is a failure as you now can't reach quorum and need to immediately get the node back up. When running 5//7/9 nodes outages are not nearly as urgent .
So why go with 8?
@@jono6379 modern voting algorithms should handle even node counts above 2 fine , though I have always built clusters in its sets out off habit.
@@jono6379 1 failure -> 7 node quorum, 2 failures -> 5 node quorum? Am I understanding this correctly?
My guess is when you have more servers, the load on each is decreased, such that occasionally one can take a backseat. So with an even number of servers, by this logic, the failure of one server would not compromise the process and it would function as normal since the “extra” would not be needed right away and thus have more time to get resolved and back up and running. Networking allows the servers to handoff tasks to neighboring servers in the network. This is also why it was mentioned in this video that “two entire servers” can go down and not cause an issue and nobody would even notice. Also, the odd number discrepancy to check for biases, that’s usually an ideal for if you have multiple tasks computing at once that put the network of servers at full load, but otherwise an even number will still work fine as one is backup.
I just decommissioned something similar at work about 2 months ago these are awesome and super reliable. The ones we decommissioned were used for high availability to radiology records in a major hospital, ran for 7 years with no issue and was only replaced for a standard life cycle upgrade of the system!
I love auctions!
So when you upgrade is that the only time it goes down?
@squidwardo7074 in our case yes. Because of the high availability of data it would take in our case 6 simultaneously failing servers to cause an outage. Or 2 simultaneously failing access switches with 2 simultaneously failing links for each server connected to each switch
Last time you lost data it was not because of lack of failsafes, but rather the fact that no one was keeping an eye on things. This may prevent failures when something goes wrong, but only if you actually respond to those failures in time. Make sure to have some sort of alert system this time around.
This, aaaaand no experiments or fancy *new* tech, it HAS TO alert someone, no use if alerting itself is down
They have actual sysadmins now fyi, like not just some of the editors or whatever happen to also be good at this stuff (such as emily for example) but like they've been hiring actual dedicated IT staff
@@mysteryboyeeThat may be, but monitoring is something that they should probably cover at some point in a video like this. Redundancy is not just about hardware. And having sysadmins does not change anything besides the fact that they would know to setup ways to monitor the situation, unless they sit on a chair on front of the server rack at all times.
The red flashing light only works until you get tired of it crying wolf, and that will probably happen before a failure
@@danielberglv259One would presume that dedicated sysadmin staff would indeed have it as part of their job description to actually do such things. They would be unfathomably stupid not to.
Thank you Jake, for calling me awesome!! :D
really enjoying the intro being back!
I REALLY want a Floatplane extra for this which dives DEEP into this stuff. I don't care if it's 3,5 hours long like the FP Exclusive of Kyle. THIS. IS. AWESOME!
I absolutely love these server oriented videos! Please never shy away from doing more! I almost wish you guys had a specialized channel for this kind of stuff and just let Jake geek out. Great stuff, LTT!
...the ending is pure gold: "..see all the lights, they never stopped blinking." :D such a powerful statement, so true for any debugging process no matter who you are!
That axle AI is crazy! Can't wait to see more historical clips showing up in new videos!
That software they use to search for videos is insane! It's pretty impressive at how accurate their searches were and how the thumbnails matched.
LOVE THE RETURN OF THE INTRO
Is returned? I guess I was watched a lot of old videos of linus 😅😅
@@ilhamkazimzadeisyourproducer I don't get it either. Isn't that the normal intro?
@@vffaa few months ago/last year supposedly their TH-cam metrics said they were losing viewer attention during the intro, so they stopped adding it in. It was mentioned on the wan show.
However personally I feel it gives the LTT a character like a proper TV show. I love it!
this is a great video to show how AI could be useful in an SMB - Small to Medium Business, not the shared drive :D I kinda wish you hadn't buried it 20 minutes into the video, but you had to, so we could see the technology that makes it reasonable to do. This is actually one of the most relevant videos of yours for me in what I do in my professional life. Good work.
THE INTRO IS BACK PEOPLE!
So are the quality viewers lol. Been drawing me back in last week or so
15:43 Jake is referring to the fact that the latency accounts for the fact that they are using DPDK (Data Plane Development Kitty) to facilitate RDMA (Remote Direct Memory Access): basically running RDMA through software, and accounting for the latency as a result (I.e, in the userspace rather than at a kennel level). This is pretty fantastic!
This is the kind of stuff that is exciting for me in IT. Although I would not go this extreme and aim for maximum value for money, it is fun to watch.
Genuinely need so much more of jake and this awesome nerdy server content.
That time jump at the beginning hooked me like a fish and I watched the whole video
I love infrastructure content and other business related IT! I could watch it all day.
Oh wow! It's an actual use for large language models and training that's ethical, reasonable, and useful! Incredible!
20:40 OK that is definitely the best feature so far, this era of information there are lots, but to actually get one that is relevant is the true power.
The camera man took my tip I am so happy :)
Jake and server upgrade video ? My favorite ngl more jake pls
This really sounds like an unsinkable ship! Good job guys!
Jinxed it! - LS
Some C-suite at Supermicro when Linus and Jake pulled the servers.
"Why do we work with these guys?"
Then again, it proves how good the setup is!
I'd love a video on the evolution of Whonnock over time. I think that's good for learning how to evolve our home server storage over time.
Fun fact: one of the Whonnock videos was the first LTT video I ever saw.
It would be kinda sick to see an educational series on IT engineering and deployment given all of the amazing professionals working on projects like this on the front and back end.
Thank you for including the intro again :D all i got to say atm :P
Jake with a full beard is a much better look for him!
Jake looked much better overall in this video. Our boi is growing up
nah facial hair looks worse on all men.
@@AliceAWilsoncan't be further from the truth
Now Linus just needs to grow one and all will be well with LMG
I love when LTT does server stuff. Keep em coming PLEASE
The concept of cluster voting was developped by Digital Equipment Corp in the 1980s for its VMS clusters. The votes and quorum mechanism did not handle just failure of a node, but more importantly prevented a node that left the cluster from continuing to function (thus proventing it from accessing drives without enforcing locks/synchrinization with the other nodes). (called cluster partitioning). In the case of a 2 node cluster, you could designate a drive as quorum disk and give it votes, so any node in the cluster that still saw that/those drives would get the votes and could then continue. (in VMS clusters, drives could either be served by a node (and accessed from other nodes via network) or be standalone with direct hardware links to each node.
(Digital developped extention to SCSI (called DSSI) to allow multiple hosts to access the same SCSI bus so multiple nodes coul write directly to the same physical drives). (later on, I beleive SCSI did get that functionality).
The freezing of a node to prevent partitioning is very clitical to prevent a node that lost its connection to the other nodes from continuing to write to disks or perform tasks without the synchronization with the other cluster nodes. (in a cluster, there was a single lock database so when a process on a node too a lock on a file (or area of file) or a resource, that lock would exist on all nodes so processes on other nodes would not get that lock until first one released it).
From a failover POV: You can have 2 nodes doing work since the locking is synchronized, so when one node fails, the other one continues and new network connectiosn all go to the second node instead of being split between the two. The other way is to have node 1 take and get lock and do all the work. Node 2 requests the lock but is put on hold because lock is unavailable. When node 1 goes down, node 2 gets the lcok and then processes all the work.
Tandem NonStop (used for mission critical stuff such as Interac/credit card) has different fault tolerance. In the same chassis, a process would run on a single CPU/RAM, but a copy would run on a designated CPU/RAM, but basically have it writes to disk/network disabled. Both processes got the same data from network/disk. So the backup was a copy of the main one, and shoudl main one go down, the backup tok over right where the backup failed since the backup had identical RAM, process state and conections to network etc.
However, while VMS clusters could be spread across multiple buildings (and up to 96 nodes), the Tandem Nonstop was within a single chassis in 1 computer room. so fault tolerant within chassis, but not disaster tolerant.
A mere smb file server just blindly executes writes/reads from any node to any area of disk/file and when you have different windows instances accessing the same time, the last one writing to it wins.
Honestly that Ai clip searching feature is the first useful, non-harmful to creators, implementation of AI I have ever seen.
"They don't know what HA means"
Me.. working in IT for 25 years: "HA means nothing if its not distributed and works as a failover cluster in a DR event"
TLDR; It's still not secure, it's not protected against floods, fires, or other disasters. It still lives in a closet. The system needs to be distributed physically before it can be called HA.
If the LTT office floods I don’t think off site high availability storage will help because LTT does not have a redundant video editing team.
@@nickfarley2268 That's not true, the team could just be moved to a temporary location with remote access to the files... that's the whole point of a DR plan. In the event his whole team dies during the flood, sure... you're right, but that would be quite a flood :)
A strong power surge could take down his whole "high availability" setup... not exactly HA.
@@oourdumb True but it also comes down to cost Vs risk this is a big calculation a business must take.
Excellent comment. Really excellent.
The serious stuff for HA is not interesting for a general audience. Is pretty boring, even for us IT professionals.
Is not "cost vs risk", it doesn't have the features for being redundant and secure. Is like building "the world's most powerfull PC" with $200, and then argue that well... it has prioritized costs...
A recomendation? Just do it right, and if it is not right for a video, don't make a video about it, period.
@@maxhennessy6676 sure, but that doesn't make it HA... that's my point. You can't say "it's a discount HA", since there is a real name for it... just a failover cluster. Even if it load balances, still not HA. HA has many requirements and physical distribution is one...
Whonnock 10
Last whonnock ever, until whonnock 11
3 years from now: How we lost all our data
So we let Elijah plug in a network cable and ...
It would be interesting to see Ceph running on this thing. CERN uses Ceph for data from the LHC.
Jake loses weight with every video we see him in. If this is intentional, keep up the good work! If it isn't, I pray you stabilize toward a healthy goal weight. ❤
12:41
I laughed way harder than i should have at that linus pillow
Now, for true redundancy, do a nightly mirror/backup over fiber optic or microwave link to an identical setup in the labs building. Call it Whonnock 10_2. God forbid another UPS or power bus catch fire in the server room. Or Linus pull 3 servers for a fail over demo by mistake.
Yep. The first thing I thought is "WHY IS IT IN THE SAME RACK?!??!"
Finally somone pointing this out, i thought i was the crazy one here
Using Ceph you can define your failure domain and run a cross datacenter cluster.
The failure domain is scalable, so right now I run a disk level failure domain at home, but in a bit I'll move to host level failure domain (once I add a few more hosts)
Not good enough..l they need backup in another part of Canada or US. That area will someday have a big earthquake off the coast and that will sever fiber optic lines and cause major power outages, etc…
Need multiple geographically diverse fiber links also, preferably to a mirror site in another city/region.
Linus: "Here it is, the final form"
_Less then 1 year later_
Linus:
I don't know why, but I forget how much networking/nas that Jake knows. When he's the one setting most of it up. Good job Jake.
19:43 first time ever hearing about Genomics Lab storage needs :).
The to go example is always video editing / visual FX, because it's more familiar of course !
Also check out Ceph I'd say. It's neat.
ever storage video I want them to check out ceph. Rook ceph is super powerful imho.
26:52 "Is this on Wi-Fi?"
-Mark 2024
🤣
Seeing the software made me really wish for someone to build a consumer storage classifier. Like just that part, you could point it to a file in your regular file system, give it somewhere to store all the generated information, and be able to manage all your pictures and videos.
Last time I checked you technically could use a business solution, but it wasn't cheap. Maybe there are cheap consumer options now though.
My mind was blown by the storage/infrastructure itself... then they started talking about that INSANE search ability, and I can't even. It's SO FREAKIN' COOL.
One thing I love about these server videos is that, unlike the high end PC builds you make, I might actually interact with one of these! Obviously not of my own, but as a professional software engineer deploying stuff for an employer. Even just making decisions based on what hardware bottlenecks will appear for different types of services! And the best part is not having to spend your own money on the outcome 😂
Obligatory ceph mention
Yeah ceph is great I have a 3 node deployment of it that's been going strong for years
yeah, given that they use TrueNAS most of they time so probably won't be going with super expensive Weka long term, or similar competition. He did mention TrueNAS cluster but TrueNas says it is based on Gluster and that support is degraded at this point. I fell Ceph would be most appropriate, and I REALLY would like to see the video for how they pull that.
Let’s get some popcorn and watch Linus drop it
Jake is the kind of employee that brings real value to an organization. Vote party for maximum raise at performance review time! 👏🏾👏🏾💯🔥
I am a small homelabber, and I love only these types of videos from LMG....I watch all of their server room/ servers/ storage kind of videos.
I first thought they were gonna use Ceph but looks like Weka is super cool too.
Weka is super expensive I hear. It sounded like a fun experiment but I bet they go with a Ceph/TrueNAS type cluster given their background.
Love the server hardware videos.. keep em coming! Supermicro web browsing, just for fun do exist
Jake is looking great. Looks like he’s been hitting the gym quite a bit
Cannot lose data… except when Linus has to be the one to replace a drive
First time hearing about Weka. Literal game changer. Very impressive.
so much energy you got there, mate.
Thanks
They got WEKA?!? I'm so jealous! I've always wanted to use this but you just can't run it for the homelab!!!! P.S. How do you backup a weka-system?
It supports snapshotting and backing up to object storage
Weka is cool but you can run ceph in a homelab
What if the building just loses power
Is this used in any big projects?
That axle AI thing is for now the most useful application of AI I’ve seen.
I have no idea about servers or storage but i still enjoy watching these
Why don't you ceph?
Day 3 of asking for Scrapyard wars 9.
Linus did say he had an idea for a scrapyard wars reboot
@@Dameworth then c'mon implement it already
Day 4 of asking Linus for his 4090
Haha, that's amazing. Congrats LTT, Supermicro, and Weka for this killer demo!
God I love the intro coming back to LTT videos! That cut from the intermission bleep to the intro had me lmao 🤣 1:38
Anyone said how good Jake is looking now!! Wasn't ever bad, but good on you bro!!!!
The capabilities of WCA are mind-blowing, and the potential for AI-assisted search is really promising.
0:55 to 1:00 - I remembered to click the like button
I love the server videos, so much crazy tech being used is amazing to watch
Jake has come a LONG way. He's REALLY impressive as a presenter.
So happy to see LTT finally investing in proper IT infrastructure, this setup is sick!