Terminal Talk

<p>Being a tester means knowing a bug's favorite hiding places. If you're a defect, you don't want to play hide and seek with Debbie Furman, a z/OS Function Tester Extraordinaire.</p> <p>We’re still looking for votes! http://ibm.biz/newstudio</p> <p> </p> <p>2:40 28 Years in Test</p> <p>3:54 Terminal Talk Anthem Right Here</p> <p>6:29 Stages of Test</p> <p>12:14 Regression Testing</p> <p>14:38 Debugging</p> <p>17:39 My Favorite Defect!</p> <p>20:11 Serialization </p> <p>20:58 "Testers are a Devious Lot"</p> <p>23:07 Managing APARS with multiple releases</p> <p>24:07 Many many many hats</p> <p>26:52 Advice for Future Testers</p> <p>30:41 Perspective</p> <p>email: contact@terminaltalk.net</p> <p>twitter: @TerminalTalk</p>

<p>Lili Tang may only have two years working in Z, but she’s already ripping through everything the platform has to offer. This episode is all about how data science works on mainframes.</p> <p> </p> <p>Fund Our New Studio!! - http://ibm.biz/newstudio</p> <p>IzODA - https://izoda.github.io</p> <p>IBM Z Technical University - Washington D.C. - https://www-01.ibm.com/events/wwe/systu/systu17dc.nsf/Registration.xsp</p> <p>0:00 A quick favor</p> <p>5:13 The daily duties of a Part-Time Data Scientist</p> <p>6:13 IzODA?</p> <p>8:16 A boatload of Data Science packages</p> <p>9:32 Marrying on-platform and off-platform data</p> <p>10:47 Yeah, but where's your terminal?</p> <p>12:27 Getting Pandas from Anacondas</p> <p>15:03 How 'bout some demos?</p> <p>17:17 Choosing the correct tools</p> <p>19:09 What about Math? I heard there's math involved</p> <p>22:36 Model Determination</p> <p>24:24 A fast-moving career</p> <p>27:11 I get analytics, buy why on Z?</p> <p>28:47 Shameless Plug Time</p>

Lili Tang may only have two years working in Z, but she’s already ripping through everything the platform has to offer. This episode is all about how data science works on mainframes.   Fund Our New Studio!! - http://ibm.biz/newstudio IzODA - https://izoda.github.io IBM Z Technical University - Washington D.C. - https://www-01.ibm.com/events/wwe/systu/systu17dc.nsf/Registration.xsp 0:00 A quick favor 5:13 The daily duties of a Part-Time Data Scientist 6:13 IzODA? 8:16 A boatload of Data Science packages 9:32 Marrying on-platform and off-platform data 10:47 Yeah, but where's your terminal? 12:27 Getting Pandas from Anacondas 15:03 How 'bout some demos? 17:17 Choosing the correct tools 19:09 What about Math? I heard there's math involved 22:36 Model Determination 24:24 A fast-moving career 27:11 I get analytics, buy why on Z? 28:47 Shameless Plug Time

From science project to enterprise hypervisor, z/VM has a multitude of uses, and a very interesting history. In this episode, Romney White walks us through why z/VM is so interesting. 0:00 Intro 2:19 Comparing VM to VMWare 3:09 From its beginning as a "science experiment" 5:20 Getting started in the VM world 6:50 "The Channel Zero Problem" 10:05 The Life and Times of VM as we know it 13:31 A Wild Unix-based operating system emerges! 16:05 Creating conditions in VM 17:56 Where is the emphasis shifting? 20:08 Why NOT run everything under z/VM? 22:06 Good Ol' CMS 24:07 The concept of Service Machines 28:55 How does one get started? 32:31 Staffing an enterprize z/VM Linux environment 34:23 Where do you see this all going? 36:36 "If it moves, we virtualize it" 37:33 Tech U Reminder   Washington D.C. Technical University: https://www-01.ibm.com/events/wwe/systu/systu17dc.nsf/Registration.xsp   contact@terminaltalk.net Twitter: @terminaltalk

<p>From science project to enterprise hypervisor, z/VM has a multitude of uses, and a very interesting history. In this episode, Romney White walks us through why z/VM is so interesting.</p> <p>0:00 Intro</p> <p><br />2:19 Comparing VM to VMWare</p> <p><br />3:09 From its beginning as a "science experiment"</p> <p><br />5:20 Getting started in the VM world</p> <p><br />6:50 "The Channel Zero Problem"</p> <p><br />10:05 The Life and Times of VM as we know it</p> <p><br />13:31 A Wild Unix-based operating system emerges!</p> <p><br />16:05 Creating conditions in VM</p> <p><br />17:56 Where is the emphasis shifting?</p> <p><br />20:08 Why NOT run everything under z/VM?</p> <p><br />22:06 Good Ol' CMS</p> <p><br />24:07 The concept of Service Machines</p> <p><br />28:55 How does one get started?</p> <p><br />32:31 Staffing an enterprize z/VM Linux environment</p> <p><br />34:23 Where do you see this all going?</p> <p><br />36:36 "If it moves, we virtualize it"</p> <p><br />37:33 Tech U Reminder</p> <p> </p> <p>Washington D.C. Technical University: <a href="https://www-01.ibm.com/events/wwe/systu/systu17dc.nsf/Registration.xsp">https://www-01.ibm.com/events/wwe/systu/systu17dc.nsf/Registration.xsp</a></p> <p> </p> <p><a href="mailto:contact@terminaltalk.net">contact@terminaltalk.net</a></p> <p>Twitter: @terminaltalk</p>

Brian Prasky knows what’s going on inside a CPU. He’s so smart he even knows what’s GOING to happen inside a CPU. This episode is all about Branch Prediction, the science and art of providing high-end CPUs with the ability to act faster based on historical data and models. If you use a computer with a processor, this is relevant to you. 1:28 What does a Logic Designer do? 4:06 The Data at Play 9:26 Building the model 13:48 How do the *others* do it? 18:23 Accuracy, Capacity, Latency 22:32 I never metacache... 23:59 How can I apply this? 25:21 Starting 'em youn 27:48 Making Progress 28:31 You ain't seen nothing yet 29:14 Breaking Apart the Chip IBM Systems Technical University @ Washington D.C. - https://www-01.ibm.com/events/wwe/systu/systu17dc.nsf/Registration.xsp Email: contact@terminaltalk.net Twitter: @terminaltalk

<p>Brian Prasky knows what’s going on inside a CPU. He’s so smart he even knows what’s GOING to happen inside a CPU. This episode is all about Branch Prediction, the science and art of providing high-end CPUs with the ability to act faster based on historical data and models. If you use a computer with a processor, this is relevant to you.</p> <p>1:28 What does a Logic Designer do?</p> <p>4:06 The Data at Play</p> <p>9:26 Building the model</p> <p>13:48 How do the *others* do it?</p> <p>18:23 Accuracy, Capacity, Latency</p> <p>22:32 I never metacache...</p> <p>23:59 How can I apply this?</p> <p>25:21 Starting 'em youn</p> <p>27:48 Making Progress</p> <p>28:31 You ain't seen nothing yet</p> <p>29:14 Breaking Apart the Chip</p> <p>IBM Systems Technical University @ Washington D.C. - <a href="https://www-01.ibm.com/events/wwe/systu/systu17dc.nsf/Registration.xsp">https://www-01.ibm.com/events/wwe/systu/systu17dc.nsf/Registration.xsp</a></p> <p>Email: <a href="mailto:contact@terminaltalk.net">contact@terminaltalk.net</a></p> <p>Twitter: @terminaltalk</p>

Peter Enrico is President of Enterprise Performance Strategies Inc, and has worked exhaustively on the mainframe platform to ensure that customers the the most from their systems. As you'll be able to tell from this interview, Peter is a natural go-to for advice on Workload Manager, Parallel Sysplex, Unix System Services, and just about anything you could hope to get your hands on. Even better, he's an an energetic speaker who loves helping demystify the platform, and we're lucky to have him on the show.   1:16 - Who are you and how did you get in here?1:50 What should everybody know about Performance?3:59 Planning for Performance6:46 Customers. Know. Their Workloads7:36 Tuning Intelligently9:05 A millisecond saved is a ... lot of money recovered11:44 Leveraging resources to maximize performance13:27 Seek and Destroy (performance sinks)15:20 Working on Performance Post-IBM15:53 How do you even begin to explain this?17:39 WLM in a nutshell20:20 ... 'cause you gotta have goals22:18 How did it all work *BEFORE* WLM?25:53 What makes z/OS Special28:41 The Power of the Access of the Data30:02 What keeps Performance exciting?32:26 What challenges do you face?37:09 Sweet sweet goodbyes Enterprise Performance Strategies - http://pivotor.comemail: contact@terminaltalk.net Twitter: @TerminalTalk

<p>Peter Enrico is President of Enterprise Performance Strategies Inc, and has worked exhaustively on the mainframe platform to ensure that customers the the most from their systems. As you'll be able to tell from this interview, Peter is a natural go-to for advice on Workload Manager, Parallel Sysplex, Unix System Services, and just about anything you could hope to get your hands on. Even better, he's an an energetic speaker who loves helping demystify the platform, and we're lucky to have him on the show.</p> <p> </p> <p>1:16 - Who are you and how did you get in here?<br /><br />1:50 What should everybody know about Performance?<br /><br />3:59 Planning for Performance<br /><br />6:46 Customers. Know. Their Workloads<br /><br />7:36 Tuning Intelligently<br /><br />9:05 A millisecond saved is a ... lot of money recovered<br /><br />11:44 Leveraging resources to maximize performance<br /><br />13:27 Seek and Destroy (performance sinks)<br /><br />15:20 Working on Performance Post-IBM<br /><br />15:53 How do you even begin to explain this?<br /><br />17:39 WLM in a nutshell<br /><br />20:20 ... 'cause you gotta have goals<br /><br />22:18 How did it all work *BEFORE* WLM?<br /><br />25:53 What makes z/OS Special<br /><br />28:41 The Power of the Access of the Data<br /><br />30:02 What keeps Performance exciting?<br /><br />32:26 What challenges do you face?<br /><br />37:09 Sweet sweet goodbyes</p> <p>Enterprise Performance Strategies - <a href="http://pivotor.com">http://pivotor.com</a><br />email: <a href="mailto:contact@terminaltalk.net">contact@terminaltalk.net</a></p> <p>Twitter: @TerminalTalk</p>

Neil Shah knows a thing or two about setting up and running z/OS systems. In this episode, he talks about how a system's Parmlib is where all the magic happens.    2:25 Meet Neil Shah2:51 Neil's got the hookup4:39 ... but stays humble5:23 So what *IS* Parmlib?6:38 Parmlib concatonations8:41 Booting up a Mainframe11:54 SYS1.IPLPARM15:10 Managing a large Parmlib17:22 Some tools to help18:08 While you're here... Linklist and Parmlib?20:53 How do you keep it all straight?24:01 Turning the Corner28:01 MY FAVORITE ACRONYM 33:35 Say goodnight, Frank   HASP - http://www.share.org/p/bl/et/blogaid=238     Email: contact@terminaltalk.nettwitter: @TerminalTalk  

<p>Neil Shah knows a thing or two about setting up and running z/OS systems. In this episode, he talks about how a system's Parmlib is where all the magic happens. </p> <p> </p> <p>2:25 Meet Neil Shah<br /><br />2:51 Neil's got the hookup<br /><br />4:39 ... but stays humble<br /><br />5:23 So what *IS* Parmlib?<br /><br />6:38 Parmlib concatonations<br /><br />8:41 Booting up a Mainframe<br /><br />11:54 SYS1.IPLPARM<br /><br />15:10 Managing a large Parmlib<br /><br />17:22 Some tools to help<br /><br />18:08 While you're here... Linklist and Parmlib?<br /><br />20:53 How do you keep it all straight?<br /><br />24:01 Turning the Corner<br /><br />28:01 MY FAVORITE ACRONYM</p> <p><br />33:35 Say goodnight, Frank</p> <p> </p> <p>HASP - <a href="http://www.share.org/p/bl/et/blogaid=238">http://www.share.org/p/bl/et/blogaid=238</a></p> <p> </p> <p> </p> <p>Email: <a href="mailto:contact@terminaltalk.net">contact@terminaltalk.net</a><br />twitter: @TerminalTalk</p> <p> </p>

What started as a discussion about Mainframe performance ended up being a rather in-depth discussion of z/OS Core Technologies, including GRS, RSM and Enque/Deque. Hey, we chase what's interesting. 1:36 Getting Started 2:42 ... and before that? 3:45 What IS RSM? 4:23 But what *IS* RSM? 6:26 The Program to LE to VSM to RSM handoff 7:15 Security and RSM 8:25 Storage Keys to the Kingdom 10:13 Better Know a Keystate 11:42 Let's talk about GRS for a bit 12:57 But really, what *IS* a resource? 13:37 ... and now we're talking Enque Deque 17:16 Don't be a GRS Goofus 19:39 Tuning shared access in Parmlib 21:03 The dangers of Hardware Reserves 22:57 To actually talk a little bit about... Performance... 26:43 Preparing to Test! 28:21 Dumpchat 30:42 Level 1, Level 2, Level What? 32:50 Performance Analysis 34:02 z/OS Tools 36:40 On the concept of Capacity Planning 40:36 Closing Remarks IBM z Systems Batch Network Analyzer (zBNA) Tool https://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS5132   Twitter: @terminaltalk Email: contact@terminaltalk.net

<p>What started as a discussion about Mainframe performance ended up being a rather in-depth discussion of z/OS Core Technologies, including GRS, RSM and Enque/Deque. Hey, we chase what's interesting.</p> <p>1:36 Getting Started</p> <p>2:42 ... and before that?</p> <p>3:45 What IS RSM?</p> <p>4:23 But what *IS* RSM?</p> <p>6:26 The Program to LE to VSM to RSM handoff</p> <p>7:15 Security and RSM</p> <p>8:25 Storage Keys to the Kingdom</p> <p>10:13 Better Know a Keystate</p> <p>11:42 Let's talk about GRS for a bit</p> <p>12:57 But really, what *IS* a resource?</p> <p>13:37 ... and now we're talking Enque Deque</p> <p>17:16 Don't be a GRS Goofus</p> <p>19:39 Tuning shared access in Parmlib</p> <p>21:03 The dangers of Hardware Reserves</p> <p>22:57 To actually talk a little bit about... Performance...</p> <p>26:43 Preparing to Test!</p> <p>28:21 Dumpchat</p> <p>30:42 Level 1, Level 2, Level What?</p> <p>32:50 Performance Analysis</p> <p>34:02 z/OS Tools</p> <p>36:40 On the concept of Capacity Planning</p> <p>40:36 Closing Remarks</p> <h4 class="smalltitle"><span style="text-decoration: underline;">IBM z Systems Batch Network Analyzer (zBNA) Tool</span></h4> <p>https://www-03.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS5132</p> <p> </p> <p>Twitter: @terminaltalk</p> <p>Email: contact@terminaltalk.net</p>

Whether you call it memory or storage, Pat Meaney is exactly the kind of guy you want in charge of it. This is the first in a series of deep-dive heavy hitter episodes about core mainframe and z/OS technologies. 1:53 A warm welcome to the show2:25 How'd You Get Started?3:51 It Gon RAIM6:52 College kids with golf carts8:33 Smart Buffer and HalfHorse Technology10:36 Books & Drawers10:36 Driving more chips without drawing more power. How do?12:19 Unplanned Outages and Epidurals14:49 DDR Revolution16:14 Comparing RAIM to RAID17:06 CSI: Mainframe Memory Division19:44 Just because you CAN do hot swaps...23:33 Getting Cosmic25:56 Gathering Call-Home Data28:16 Variable... bandwidth? 29:39 On Golden Copy30:32 Optimized for Mixed Workloads31:03 If This Then That... on chip33:49 The Bag Test34:30 Maximizing Cooling in an ever-shrinking area36:48 How NOT to leave your mark on mainframe design37:24 Finding performance gains in everything40:01 The Next Steps in Memory41:39 Future Memories... this weekend at the Hayden Planetarium. Twitter: @terminaltalkEmail: contact@terminaltalk.net  

<p>Whether you call it memory or storage, Pat Meaney is exactly the kind of guy you want in charge of it. This is the first in a series of deep-dive heavy hitter episodes about core mainframe and z/OS technologies.</p> <p><br />1:53 A warm welcome to the show<br />2:25 How'd You Get Started?<br />3:51 It Gon RAIM<br />6:52 College kids with golf carts<br />8:33 Smart Buffer and HalfHorse Technology<br />10:36 Books & Drawers<br />10:36 Driving more chips without drawing more power. How do?<br />12:19 Unplanned Outages and Epidurals<br />14:49 DDR Revolution<br />16:14 Comparing RAIM to RAID<br />17:06 CSI: Mainframe Memory Division<br />19:44 Just because you CAN do hot swaps...<br />23:33 Getting Cosmic<br />25:56 Gathering Call-Home Data<br />28:16 Variable... bandwidth? <br />29:39 On Golden Copy<br />30:32 Optimized for Mixed Workloads<br />31:03 If This Then That... on chip<br />33:49 The Bag Test<br />34:30 Maximizing Cooling in an ever-shrinking area<br />36:48 How NOT to leave your mark on mainframe design<br />37:24 Finding performance gains in everything<br />40:01 The Next Steps in Memory<br />41:39 Future Memories... this weekend at the Hayden Planetarium.</p> <p>Twitter: @terminaltalk<br />Email: contact@terminaltalk.net</p> <p> </p>