I am observing something that I cannot explain. I am trying to come up with a good filesystem configuration using HP P420i controller and SSDs. Out of curiosity I have tried BTRFS (still unstable so I can't really expect to be able to use it) and noticed that the read speed is about 150% of ext4 ...
I am only interested in part of those output lines
\\130.141.90.70\Sofon$\SOFONDB\SOFON_Logfiles\EPALQueueBackground\EPALQueueBackground_log.txt:1855:8/27/2014 10:14 AM EPAL: xml itemdef start generation for Product version (860315 - PageWriter TC70 Cardiograph - 6 - 0) \\130.141.90.70\Sofon$\SOFONDB\SOFON_Logfiles\EPALQueueBackground\EPALQueueBackground_log.txt:1856:8/27/2014 10:14 AM EPAL: xml itemdef generated for Product version (**860315 - PageWriter TC70 Cardiograph** - 6 - 0)
@JourneymanGeek Yesterday I found out the forensic lessons I'm having are taking place on a building that's the backbone for all the city's government institutions
Moving a folder from local linux machine through ssh how come this doesnt work?
scp /var/www/system_learningbank/ root@89.104.220.207:/var/www/system_learningbank/ root@89.104.220.207's password: bash: scp: command not found lost connection
@Bob: eh. If I was building a pure VM box now, I'd go with a core i5 with VT-D
Actually I went with a core i7 because I was under budget and I figured having vt-d would mean I could just stick more ram, and use it just to run VMs when I next upgraded
@Bob Dropbox, Crashplan, Outlook, console windows, task manager, skype, Chrome with a bazillion tabs, and whatever other background processes I have easily prevent the CPU from dropping below 10% utilization
@DarthAndroid exactly -- my issue isn't so much that a single task can peg the CPU (though during loading of many programs the CPU can be pegged for several seconds, so load times can still be reduced by higher single-thread perf) but that you have to shut down a lot of background stuff when you're on a low-end system to keep it sane
@DarthAndroid 1. That's very different from maxing it out, 2. If those listed processes are causing significant constant utilisation, something else is wrong.
@JourneymanGeek (assuming you mean CPU cache) basically, instead of memory throughput being a potential bottleneck, that bottleneck is replaced with CPU cache bottleneck, which is a much, much higher ceiling
memory bandwidth and/or latency is pretty damn fast on, say, DDR3-2000 or so, and it'll be amazing with DDR4 coming out in the next few years, but cache on the CPU package is still amazingly faster than RAM
and having lots of data that you're working with frequently within the CPU cache prevents those requests from ever hitting your system memory
If you're switching a lot between different programs in general use, it's not gonna make much difference. If you're doing some high-performance computing task, it'll matter more.
CPU cache is awesome for stuff like, having multi-hundred-megabyte data structures (trees, hash tables, arrays, etc) in RAM and having a program run through them processing items repeatedly in a loop with an O(n) or worse algorithm
a lot of programs we use on a daily basis have algorithms like those within them, but we often end up only processing smallish amounts of data, not many thousands or millions of items
@Bob To give a better idea of what I mean by constant utilization, at any given time there are 20+ processes using on average 0.5% CPU. These 20+ processes are different every second, but there are at least 20 of them each second
@DarthAndroid I still think you have something weird going on. I haven't seen my process count drop below 150 in years, and it's often above 200. None of them are that hungry.
@Bob Crashplan uses a fair bit of CPU; I use it so I know :P even when you have nothing going on it is periodically scanning for any changes to files it needs to back up
@Bob I see it on both desktop and work laptop, never seemed odd to me. The processor is speed-stepping, so it's often at reduced speeds, but there's still always something going on.
That's way faster than what an SSD would be capable of, even with eSATA version 3
I've occasionally applied enough memory pressure on this system (the one I'm typing on now with 2 GB of DDR3-1333 and the page file entirely on the Samsung 850 Pro attached to eSATA 3 Gbit/s) to cause severe swapping... and when I try to restore, say, an Eclipse-based application, I see black widgets (the textures have been swapped out to disk) for about 1/2 of a second.
When the pagefile is on the HDD, I don't even see the window itself for 3 or 4 seconds, and then the textures take about 9 seconds to fully load.
Eh. It's an improvement.
12 seconds down to 0.5? 8 GB of RAM still runs circles around it, but the SSD in turn runs circles around the HDD.
When the away team prepares to beam down, they invariably assume the transporter formation: standing upright, usually with arms slack aside, all facing the camera. When they beam down, they reappear in the same formation and poses, as that's how the transporter works.
However, they do this even ...
geez, 7200rpm disks are slow enough... unfortunately, they do drain a significant amount of power, and it's hard to design an HDD that can reliably and quickly do many start->stop cycles, so you have to keep it spinning constantly, which drains energy :/
my next laptop will definitely have an SSD, I can say that for certain... you just have to deal with the expense and/or write cycles and/or capacity... there are way too many downsides to HDDs in laptops. shock damage, constant power drain, slooooooow, etc
you can make HDDs work on a desktop with the right RAID controller ;) and energy is less important
you might want to buy your next laptop with the cheapest-available storage solution (just because they require you to have one -_-) then buy a Samsung 850 Pro to stick in there
@Bob grabbing the entire disk at a block level will do it... other than that, you'd have to grab every partition, since the EFI SYSTEM PARTITION is just a regular partition, there's nothing in the "MBR" (the GPT equivalent) except a list of what partitions there are and what are bootable
there's no actual boot code in the GPT outside of what's within the partitions themselves, so if you can't for some reason clone the entire disk at a block level, taking all the partitions should be fine
well like I said the GPT only keeps track of flags and the list of partitions and their start and end locations in logical blockspace (and their optional "label")
copying the partitions as they are, plus their flags and labels, should be fine
good thing you aren't booting Linux by GUID though because copying over your partition would probably change its GUID
for Windows, worst case, you just have to use bcdedit to re-do the boot code in the EFI SYSTEM PARTITION
which can be done from the recovery console on an install DVD
> Based on the nature of this issue, HP is requesting that you immediately discontinue use of the subject AC power cords by removing it from the wall outlet and ordering a replacement AC Power Cord upon verification from the program Web site. NOTE:The notebook will function using battery power.
Yes, because my laptop will run long enough for the delivery...
> Your AC Power Cord is not affected and is not part of this replacement program. You may continue using your AC Power Cord. Thank you for your cooperation.
as for that pc parts picker link you provided earlier, Bob, I think that guy isn't exactly going overboard, if he plans to play AAA games going forward. the 4790K isn't even the highest-end CPU available; he isn't going for the enthusiast platform! so you can give him credit for having restraint in that matter :P
> The system is not designed to do what you are asking it to do. If a SATA hard drive is installed the system will not look past that drive in the boot order. So when you have the mSSD and the SATA drive in there it is trying to boot only from the SATA drive, which is not bootable. The only way to have the mSATA drive be the Windows boot drive is to install a "bootloader" on the SATA drive, which starts the boot process and then hands off to the mSATA drive to complete the sequence.
in the section of wikipedia on smilis sadis about reproduction, it says "None." and moves on to the next topic about how they always complain of their low salary and limited buying power
> If anyone uses SharePoint at work and knows what it looks like out of the box and is curious what it can turn into you can find a list of public websites that run on SharePoint here.
> To view this content, please install Microsoft Silverlight
> RAPID is a software-based feature in Samsung's most recent build of Magician. Simply, it uses a gigabyte of system memory for caching hot data. Frequently-used applications are stored in RAM, ideally yielding much faster accesses when that data is needed over and over. Unlike other RAM-based caching solutions, RAPID keeps cached data persistent between reboots by writing information to the SSD itself.