<?xml version="1.0" encoding="utf-8"?>
<rss xmlns:atom="http://www.w3.org/2005/Atom" version="2.0"><channel><title>Disqus - Latest Comments for DeepStorage</title><link>http://disqus.com/by/DeepStorage/</link><description></description><atom:link href="http://disqus.com/DeepStorage/comments.rss" rel="self"></atom:link><language>en</language><lastBuildDate>Thu, 18 Aug 2016 21:35:42 -0000</lastBuildDate><item><title>Re: Seagate Ups The Ante with 60TB SSD</title><link>https://blog.architecting.it/seagate-ups-the-ante-with-60tb-ssd/#comment-2845218338</link><description>&lt;p&gt;They're for the webscalers.  If you're facebook your lukewarm data is something like 1-2 IOPS/GB more than spinning disks can do cost effectively but that comes out to around 60K IOPS for this drive.  Put 60 of these in a 4U JBOD and it's perfect for that use case.   They'll sell a truckload each to 10 customers.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Thu, 18 Aug 2016 21:35:42 -0000</pubDate></item><item><title>Re: Vote No on VMUG Board</title><link>http://www.deepstorage.net/NEW/vote-no-on-vmug-board/#comment-2248493976</link><description>&lt;p&gt;Interesting that Article VI 1.D.e says "Write-in candidates may be accepted according to VMUG's BGPs." but no write in candidates are allowed.&lt;/p&gt;&lt;p&gt;Of course those referenced BGPs aren't anywhere easy to find on the VMUG site either.  I'm just a geek so BGP is Boundary Gateway Protocol to me.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Fri, 11 Sep 2015 14:06:18 -0000</pubDate></item><item><title>Re: Idaho city's ordinance tells pastors to marry gays or go to jail</title><link>http://www.washingtontimes.com/news/2014/oct/20/idaho-citys-ordinance-tells-pastors-to-marry-gays-/#comment-2107781658</link><description>&lt;p&gt;That argument was made about serving blacks in restaurants in the 1960s. It failed.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Mon, 29 Jun 2015 21:49:53 -0000</pubDate></item><item><title>Re: It Came From R&amp;D: 7 Vintage FrankenPCs</title><link>http://www.pcmag.com/slideshow/story/335346/it-came-from-r-d-7-vintage-frankenpcs#comment-2088805409</link><description>&lt;p&gt;My partner at the time and I wrote the CP/M BIOS for the SpectraVideo SV318 and SV328 (more power no joystick).&lt;/p&gt;&lt;p&gt;The weirdest computer I ever had in house was a Canon PC that combined the phone (handset hook on the side), answering machine (with animated Cassette when you played back, and FAX (with a thermal printer from a roll that came out the top).  8086 based circa 1985-86&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Fri, 19 Jun 2015 18:30:07 -0000</pubDate></item><item><title>Re: Faster Ethernet Gets Weird</title><link>https://blog.fosketts.net/2015/06/19/faster-ethernet-gets-weird/#comment-2088546662</link><description>&lt;p&gt;To further the thought over the years Ethernet has also eliminated most other comm channels with broader use cases come the need for more variants like NBASE-T for fast access points over twisted pair with POE but we aren't discarding the slow channel versions (1Gbase-T ports almost always still support 10Mbps).&lt;/p&gt;&lt;p&gt;One minor correction/addition to the vendor mix. While Avago did buy Broadcom Qlogic bought Broadcom's Ethernet adapter chip business earlier. That leaves Avago (Emulex), Intel and Qlogic (Broadcom) as the volume LOM chip vendors.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Fri, 19 Jun 2015 16:02:19 -0000</pubDate></item><item><title>Re: We Can&amp;#8217;t Let John Deere Destroy the Very Idea of Ownership</title><link>http://www.wired.com/2015/04/dmca-ownership-john-deere/#comment-1981324264</link><description>&lt;p&gt;Wait until GM or Deere decides to change the (not fully disclosed to anyone who isn't a lawyer specifically looking for them)  terms of the license for the engine management software to make the software, like PC software is, non-transferable.  Now GM won't have to compete with used cars because when you sell your car they revoke the software license bricking it.  GM could charge $3000 license transfer fees.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Wed, 22 Apr 2015 10:40:16 -0000</pubDate></item><item><title>Re: Viewing Clintonemail.com as Guerrilla IT</title><link>http://www.deepstorage.net/NEW/viewing-clintonemail-com-as-guerrilla-it/#comment-1947795999</link><description>&lt;p&gt;Like all guerrilla IT Mrs. Clinton's decision to run her own server create security issues.  However given the poor state of IT security and archiving at State those seem to be about in line with what would have happened if she had used the State Department servers.&lt;/p&gt;&lt;p&gt;My main point, which most commenting, especially on LinkedIn, have missed or ignored in their need to attack or defend Mrs. Clinton, is that Mrs. Clinton using her personal email server is not, except for the fact that Mrs. Clinton is a polarizing political figure, substiantially different than a new VP of sales using his/her personal Dropbox account to store corporate data.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Sun, 05 Apr 2015 15:03:28 -0000</pubDate></item><item><title>Re: Watch: A Professor&amp;#8217;s Amazing Idea On Dealing With Open Carry Activists</title><link>https://www.alancolmes.com/2014/07/26/watch-a-professors-amazing-idea-on-dealing-with-open-carry-activists/#comment-1520060456</link><description>&lt;p&gt;we all have a constitutional right to free speech. Should I exorcise it all the time by cursing at the top of my lungs or telling the open carry morons that they must have such a small penis to have to show off their steel substitute all the time?&lt;/p&gt;&lt;p&gt;No that would be impolite, uncivilized and unreasonable. Almost as impolite, irresponsible, unreasonable and uncivilized as walking into a family restaurant with long guns and scaring the crap out of people.&lt;/p&gt;&lt;p&gt;As we should have learned from Spiderman with rights come responsibilities. These morons want the former without the later proving they didn't deserve the former in the first place.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Fri, 01 Aug 2014 13:57:06 -0000</pubDate></item><item><title>Re: What&amp;#8217;s Wrong with Tape?</title><link>http://archive.architecting.it/2012/10/26/whats-wrong-with-tape/#comment-699102485</link><description>&lt;p&gt;You're conflating a couple of things Chris.  Because today's disks (basically everything since that ST-506 including PATA, SATA, SAS and SCSI) combines media, drive and controller in a single package the physical side of the format is settled in the one package. On the other hand the file system and RAID layout of the disk are different for each system you put that drive into.&lt;/p&gt;&lt;p&gt;If you take a disk drive off the shelf you have to guess what file system the disk was formatted in.  If you take an EXT2 disk and stick it in a PC running Windows it's just as greek as if you try to read a Networker tape with Backup Exec. (BE can read old Arcserve tapes just as a Mac can read NTFS)&lt;/p&gt;&lt;p&gt;Think about the difficulties of having a RAID set of 3.5" disks from a Celerra and reading them on a NetApp?&lt;/p&gt;&lt;p&gt;Snice the world standardized on LTO life has gotten a lot simpler.  HP and IBM still make LTO-3 drives so every LTO tape ever written can be read with NEW drives.&lt;/p&gt;&lt;p&gt;If we have a need for removable storage, and services like Amazon's Glacier make that questionable for many people, we're going to have to track where the media came from and where it can go back to.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Fri, 02 Nov 2012 18:09:48 -0000</pubDate></item><item><title>Re: What&amp;#8217;s Wrong with Tape?</title><link>http://archive.architecting.it/2012/10/26/whats-wrong-with-tape/#comment-699035989</link><description>&lt;p&gt;No you couldn't take that ST-506 disk from your 386 and put it in a modern PC with only a SATA interface. You'd need an old DLT7000 drive to read a DLT7000 tape but at least that would be SCSI and you can get a PCIe SCSI HBA.&lt;/p&gt;&lt;p&gt;No one ever made PCI or PCIe ST-506 controllers so you'll need an old computer with an ISA bus slot for the disk controller and then need to enter the drive geometry (number of heads, tracks, sectors per track and the interleave used when  the data was written) into the old version of DOS (or Windows NT 4.0 or earlier) to read the disk.&lt;/p&gt;&lt;p&gt;LTFS is a big step up from TAR but the backup vendors have reasons, granted many about customer control, for proprietary formats.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Fri, 02 Nov 2012 17:10:12 -0000</pubDate></item><item><title>Re: Reminiscing About the Cubix ERS, My First Bladed Server System</title><link>http://blog.fosketts.net/2012/02/21/cubix-ers-blade-server-system/#comment-444842830</link><description>&lt;p&gt;Cubix and the proto-blade had been around a while by the time you saw them my young friend.  Back in the early '80s Industrial Micro Systems, Cubix's original name, and several other vendors were selling Z-80 based systems using proto-blades on the S-100 bus.  &lt;/p&gt;&lt;p&gt;One processor, the master, controlled access to shared I/O resources, primarily disk drives both floppy and hard, while up to 16 slave cards with processor, memory and serial ports ran applications.  Generally each slave acted as a single user's PC and the users had dumb, or VT-100 not so dumb, terminals on their desks.  &lt;/p&gt;&lt;p&gt;The whole thing was managed with either Digital Research's CP-Net or a better CP/M compatible OS TurboDOS. The master acted as a disk server, like 3Com's 3Share before they switched to LANmanager, so shared access to data required applications to be written to the system there being no file or record locking semantics.  &lt;/p&gt;&lt;p&gt;As the PC took over Cubix and J&amp;amp;L (which became Chatcom) moved to x86 DOS architectures and sold most of their systems for remote access as each "blade" could run a modem for PC Anywhere type connections.&lt;/p&gt;&lt;p&gt;Curious facts: &lt;/p&gt;&lt;p&gt;1 - My brother wrote, and my company at the time Pro-Comp Systems published, the definitive (ok only) book on TurboDOS, TurboDOS made Easy.&lt;/p&gt;&lt;p&gt;2 - Novell Data Systems sold CP-Net based systems using RS-422 to link the master to the slaves before Ray Noorda came around and hired SuperSet to create NetWare.&lt;/p&gt;&lt;p&gt;3 - I'm really old cause I remember all this.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Tue, 21 Feb 2012 10:52:54 -0000</pubDate></item><item><title>Re: Virtualization Field Day 2 &amp;#8211; Silicon Valley</title><link>http://blog.fosketts.net/2012/02/15/virtualization-field-day-2-silicon-valley/#comment-440327717</link><description>&lt;p&gt;Many congratulations to the proud papa.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Wed, 15 Feb 2012 19:22:20 -0000</pubDate></item><item><title>Re: Storage History: The 3Server</title><link>http://blog.fosketts.net/2007/06/25/storage-history-the-3server/#comment-63238490</link><description>&lt;p&gt;Actually the 3Server with 3+Share wasn't a file server but what we would now call a SAN array.  Workstations sent head, disk, sector (SCSI was new then so LBA not a standard) requests.  Volumes were R/W for a single user or R/O for all.  Shared R/W required applications to support APIs for access.&lt;/p&gt;&lt;p&gt;That's why Novell killed them. File services much better idea for user workstations.&lt;/p&gt;&lt;p&gt;3+Open used SMB and was a file server.&lt;/p&gt;&lt;p&gt;SMB first appeared in IBM LAN Program/DOS 3.1 and MS OEMed it as MS-Net.  DOS was just too limited as a server OS so they partnered with 3Com to add networking to OS/2 in LAN Manager.  3Com got hosed on the deal by the way.&lt;br&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Mon, 19 Jul 2010 15:35:36 -0000</pubDate></item><item><title>Re: Can You Really Build an All-Ethernet Data Center with FCoE?</title><link>http://wikibon.org/blog/all-ethernet-data-center-fcoe/#comment-54780122</link><description>&lt;p&gt;Stu,&lt;br&gt;Brocade has an FCoE blade for their DCX director (24 ports) so you can connect FCoE to the core.&lt;/p&gt;&lt;p&gt;On the other hand Greg has a point calling it proprietary.   There are 2 switch vendors and the DCB standards haven't been adopted yet.  It will be a real ecosystem when the Ethernet vendors that aren't part of the FC oligopoly join in.&lt;/p&gt;&lt;p&gt;&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Fri, 04 Jun 2010 16:26:56 -0000</pubDate></item><item><title>Re: How Help Desk SLA&amp;#8217;s lie &amp;#8211; &amp;#8220;Four Hour Response&amp;#8221; my butt.</title><link>http://etherealmind.com/opinion-help-desk-sla-are-all-lies/#comment-221252843</link><description>&lt;p&gt;Remember this applies to your hardware support too.  They have to send a tech in 4 hours but he doesn't have to bring the PARTS.&lt;/p&gt;&lt;p&gt;I once had Compaq tell me a new SCSI backplane for a server was 4 weeks!  A new server 2 days.&lt;/p&gt;&lt;p&gt;So I bought the new server, swapped backplanes and sent the new server back telling AP not to pay.&lt;/p&gt;&lt;p&gt;Took a bit of yelling to resolve that one.&lt;/p&gt;&lt;p&gt;Now I prefer to have spares.&lt;/p&gt;&lt;p&gt;and don't get me started with the tech that said that my swapping out a failed hard drive for a spare voided my contract.&lt;/p&gt;&lt;p&gt;DeepStoragenet&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Sun, 07 Mar 2010 16:40:27 -0000</pubDate></item><item><title>Re: 5 Tips To Avoid Violating NDAs</title><link>http://blog.fosketts.net/2010/02/22/5-tips-avoid-violating-ndas/#comment-35870430</link><description>&lt;p&gt;One key to your point 4 is that a well written NDA clearly says if the information has become public some other way the NDA is no longer in force.  I've been know to cut and paste (sometimes literally w/Rubber Cement and everything) that clause in before signing.&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Mon, 22 Feb 2010 11:55:02 -0000</pubDate></item><item><title>Re: What Is The Secret To Efficient Hard Disk Drives?</title><link>http://blog.fosketts.net/2009/08/25/efficient-disk-drives/#comment-15426516</link><description>&lt;p&gt;We should all remember that faster rotational speeds are more about speeding random I/O where you have on average 1/2 a rotation to wait for your data after each head movement than sequential I/O.  Since the apps we're likely to use 1.5TB or larger drives in are more sequential (OK dedupe breaks that somewhat) than random 5400RPM probably isn't much of a performance hit.&lt;/p&gt;&lt;p&gt;Now I have to test. Calling WD.&lt;/p&gt;&lt;p&gt; - Howard&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Wed, 26 Aug 2009 12:47:05 -0000</pubDate></item><item><title>Re: Cisco Launching Full Assault On Data Center Server Market</title><link>http://gestaltit.com/featured/top/stephen/cisco-assault-data-center-server/#comment-11584300</link><description>&lt;p&gt;I find it amusing that the same members of the chattering class, excluding of course thee and me, that wrote a year ago how Dell isn't a serious player in the server market because they only have x86 servers with 1-4 sockets say Cisco's UCS servers are the best thing since sliced bread.  From where I sit on my substantial ass Cisco is shipping next year's servers this year.  They've added more memory capacity, although using it slows memory access slightly, and made 10GgigE standard where IBM, HP Dell Etc. have it as an option.  Of course in those organizations where the network guys have more clout than the server guys that will be enough.&lt;/p&gt;&lt;p&gt; - Howard&lt;/p&gt;</description><dc:creator xmlns:dc="http://purl.org/dc/elements/1.1/">Howard Marks</dc:creator><pubDate>Mon, 22 Jun 2009 19:32:08 -0000</pubDate></item></channel></rss>