From awd at cs.cmu.edu  Tue May  2 18:16:21 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Tue, 2 May 2023 18:16:21 -0400
Subject: Thesis Defense - May 3, 2023 - Kin Gutierrez Olivares - Applied
 Mathematics of the Future
In-Reply-To: <CAJvAoytx6PFiDc-wP9gqbbZXb1Q_4tW460VjrN_8GY3bGrqgoA@mail.gmail.com>
References: <2235987c-7e5a-7569-3b3e-d621e213c8e9@andrew.cmu.edu>
 <CAJvAoytx6PFiDc-wP9gqbbZXb1Q_4tW460VjrN_8GY3bGrqgoA@mail.gmail.com>
Message-ID: <CAJvAoys_vCFxgGYye36EoYY8wnMRpyiL34jP8ToCNFhj35YDKg@mail.gmail.com>

After an intellectual feast we were provided by Chirag today in his thesis
defense, we are going to double the load tomorrow with Kin's defense!

Just a reminder...

See you all there tomorrow (on zoom only).

Cheers,
Artur

PS This major event frequency matches the previous record of the Auton Lab:
producing two doctors in one week :)

On Mon, Apr 17, 2023 at 4:33?PM Artur Dubrawski <awd at cs.cmu.edu> wrote:

> Team, please mark your calendars for this intellectual feast brought to us
> generously by Kin.
>
> Cheers,
> Artur
>
> PS The first week of May will be loaded with events for us. Expect at
> least one more phd thesis defense plus at least two new thesis proposals
> coming out of our Lab that week!
>
>
> ---------- Forwarded message ---------
> From: Diane Stidle <stidle at andrew.cmu.edu>
> Date: Mon, Apr 17, 2023, 3:41 PM
> Subject: Thesis Defense - May 3, 2023 - Kin Gutierrez Olivares - Applied
> Mathematics of the Future
> To: ml-seminar at cs.cmu.edu <ML-SEMINAR at cs.cmu.edu>, <
> stine at wharton.upenn.edu>, <robertastine at icloud.com>
>
>
> *Thesis Defense*
>
> Date: May 3, 2023
> Time: 12:30pm (EST) - Remote only presentation
> PhD Candidate: Kin Gutierrez Olivares
>
> *Title: **Applied Mathematics of the Future*
>
> Abstract:
> Novel learning algorithms have enhanced our ability to acquire knowledge
> solely from past observations of single events to learn from the
> observations of several related events. This ability to leverage shared
> useful information across time series is causing a paradigm shift in the
> time-series forecasting practice. Still, machine learning-based forecasting
> faces pressing challenges that limit its usability, usefulness, and
> attainable real-world impact, including human interpretability, the ability
> to leverage structured information, generalization capabilities, and
> computational costs. This thesis tackles these challenges by bridging the
> gap between machine learning and classic statistical forecasting methods.
>
> We organized the thesis as follows. We introduce the time-series
> forecasting task, accompanied by a short review of modern forecasting
> models, their optimization, and forecast evaluation methods. In the
> following chapters, we present our approach with three case studies. First,
> we augment state-of-the-art neural forecasting algorithms with
> interpretability capabilities inspired by time series decomposition
> analysis; we illustrate its application in the short-term electricity price
> forecasting task. Second, we improve neural forecasting generalization and
> computational efficiency in the long-horizon setting through a novel
> wavelet-inspired algorithm that assembles its predictions sequentially,
> emphasizing components with different frequencies and scales. Third, we
> tackle the hierarchical forecasting task, a regression problem with linear
> aggregation constraints, by augmenting neural forecasting architectures
> with a specialized probability mixture capable of incorporating the
> aggregation constraints in its construction. Our approach improves upon the
> current state-of-the-art in each of the considered domains.
>
> *Thesis Committee:*
> Artur Dubrawski (Chair)
> Barnabas Poczos
> Russ Salakhutdinov
> Robert A. Stine (University of Pennsylvania)
>
> Link to the draft document:
>
> https://drive.google.com/drive/folders/1PZX-1cFVy83M5McOL16WPHIUJYeHEcxn?usp=share_link
>
> Zoom meeting link:
> https://cmu.zoom.us/j/96712150727?pwd=UXJIOGorbXZjbEtKdWVPdEFTV3ZPUT09
>
> --
> Diane Stidle
> PhD Program Manager
> Machine Learning Department
> Carnegie Mellon Universitystidle at andrew.cmu.edu
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230502/c3fb1dcb/attachment.html>

From awd at cs.cmu.edu  Tue May  2 18:19:43 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Tue, 2 May 2023 18:19:43 -0400
Subject: REMINDER: Thesis Proposal - May 3, 2023 - Ifigeneia Apostolopoulou -
 Decision-Making Under Latent Factors
In-Reply-To: <CAJvAoyt2FQx5=33=cWpmGR40eZ3WZwBHBo+QmeBc9qhGQwF59w@mail.gmail.com>
References: <70e8e316-9d50-e4c3-c4f9-508e1f73cac1@andrew.cmu.edu>
 <CAJvAoyt2FQx5=33=cWpmGR40eZ3WZwBHBo+QmeBc9qhGQwF59w@mail.gmail.com>
Message-ID: <CAJvAoysqybG16gbM+gdr33fH45SnWfxodg-s9aK_zwiai2NPgg@mail.gmail.com>

Another reminder about another milestone talk.

Ifi, tomorrow late afternoon.

Zoom only.

See you all there.

Cheers,
Artur

On Sat, Apr 29, 2023 at 9:44?AM Artur Dubrawski <awd at cs.cmu.edu> wrote:

> And, to provide icing on the cake of multiple excellent Auton
> presentations next week, please mark your calendars for Ifi's doctoral
> thesis proposal talk on Wednesday.
>
> Cheers
> Artur
>
> ---------- Forwarded message ---------
> From: Diane Stidle <stidle at andrew.cmu.edu>
> Date: Fri, Apr 28, 2023 at 2:38?PM
> Subject: Thesis Proposal - May 3, 2023 - Ifigeneia Apostolopoulou -
> Decision-Making Under Latent Factors
> To: ml-seminar at cs.cmu.edu <ML-SEMINAR at cs.cmu.edu>, <fusi at microsoft.com>, <
> doucet at stats.ox.ac.uk>, <arnaud-doucet at google.com>
>
>
> *Thesis Proposal*
>
> Date: May 3, 2023
> Time: 3:30pm (EST) (Remote only)
> Speaker: Ifigeneia Apostolopoulou
>
> *Title: **Decision-Making Under Latent Factors*
>
> Abstract:
> Complex data arise in many fields including health care, neuroscience,
> computer vision, and many others. However, they often exhibit simple, yet
> unobserved patterns. These patterns can be represented by latent variables
> that augment the observations in the system of interest. This thesis
> surveys how to use latent variable models to guide decision-making. We
> approach decision-making as a workflow of three cognitive tasks: i)
> Representation Learning, ii) Temporal Modeling, iii) Uncertainty-Aware
> Reasoning. We propose that the processing ability required for each task
> can be improved by the use of latent variable models. To demonstrate that,
> we first discuss the types of patterns that latent variables need to
> capture for each task. Further, we develop inference tools that allow for
> expressive posterior distributions. Finally, our empirical analysis
> demonstrates the superior performance of latent variable models on several
> machine learning problems compared to their counterparts that operate
> purely in the observed data space.
>
> *Thesis Committee: *
> Artur Dubrawski (Chair)
> Ruslan Salakhutdinov
> Tom Mitchell
> Nicolo Fusi   (Microsoft)
> Arnaud Doucet (Google Deepmind)
>
> Zoom Link:
> https://cmu.zoom.us/j/96935140004?pwd=akNlbmU4TnRPWWlmWE5wWGIxaU5MZz09
>
>
> --
> Diane Stidle
> PhD Program Manager
> Machine Learning Department
> Carnegie Mellon Universitystidle at andrew.cmu.edu
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230502/a9ba6eca/attachment.html>

From predragp at andrew.cmu.edu  Wed May  3 08:40:23 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Wed, 3 May 2023 08:40:23 -0400
Subject: Fwd: SCS Computing: SCS Database Maintenance Outage 05/15/2023
In-Reply-To: <64524e95203a2_1ad742aad9f65e5c426415@massmail-02.andrew.cmu.edu.mail>
References: <64524e95203a2_1ad742aad9f65e5c426415@massmail-02.andrew.cmu.edu.mail>
Message-ID: <CAEFFt-G796z+zKD0vjzszn-43e17UxYT_YxxzHmZ3NJV3COXOg@mail.gmail.com>

---------- Forwarded message ---------
From: <help at cs.cmu.edu>
Date: Wed, May 3, 2023, 8:15 AM
Subject: SCS Computing: SCS Database Maintenance Outage 05/15/2023
To: <scs-all at cs.cmu.edu>


*Day/Date:* Monday, May 15, 2023

*Time:* 6:00PM - 10:00PM

*Service Affected:* RAMS and the Website Redirect Databases

*Details:*

On Monday, May 15, 2023, SCS Computing Facilities will perform scheduled
maintenance on the RAMS Oracle Database and the Website Redirect Database
which support a variety of SCS infrastructure services and costing
functionality. During this maintenance outage the following items will not
be available:

   - CASE System (Configuration and Administration of SCS Equipment)
   - Web Reports
   - Procurement Application
   - Poster Printing Submission Application
   - Hostinfo utility

*Additionally:*

   - Changes to system, network and charging configurations for equipment
   will not be available.
   - SCS websites redirecting to external sources will experience a brief
   outage.

To verify the authenticity of this message, visit the SCS Computing
Facilities page at
https://computing.cs.cmu.edu/dashboard/2023/scs-db-maint-05-15-2023.

Please contact the SCS Help Desk at 412-268-4231 or by sending mail to
help at cs.cmu.edu with any questions or concerns regarding this maintenance
period.

Thank you,

SCS Computing Facilities

https://computing.cs.cmu.edu


_______________________________________________
ri-people mailing list
ri-people at lists.andrew.cmu.edu
https://lists.andrew.cmu.edu/mailman/listinfo/ri-people
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230503/dac533c7/attachment-0001.html>

From predragp at andrew.cmu.edu  Thu May  4 19:23:24 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Thu, 4 May 2023 19:23:24 -0400
Subject: GPU11
In-Reply-To: <CAN_-bZJ8i_KrZo-OP9drfR5_EfAthwrJoL6OMvUZS_nv=L1+PQ@mail.gmail.com>
References: <CAN_-bZJ8i_KrZo-OP9drfR5_EfAthwrJoL6OMvUZS_nv=L1+PQ@mail.gmail.com>
Message-ID: <CAEFFt-GxBEAB-J8WmA+nqxCPFtA7wy5_7PxsC0-J40fa3c4=Ng@mail.gmail.com>

The HDD died. Luckily, I was smart enough to leave the second unused
storage drive in that computing node.

root at gpu11$ lsblk
NAME              MAJ:MIN RM   SIZE RO TYPE MOUNTPOINT
sda                 8:0    0   1.8T  0 disk
??sda1              8:1    0   1.8T  0 part
sdb                 8:16   0   1.8T  0 disk
??sdb1              8:17   0   1.8T  0 part /home/scratch2
nvme0n1           259:0    0 238.5G  0 disk
??nvme0n1p1       259:1    0     1G  0 part /boot
??nvme0n1p2       259:2    0 237.5G  0 part
  ??sl_gpu11-root 253:0    0    50G  0 lvm  /
  ??sl_gpu11-swap 253:1    0     4G  0 lvm  [SWAP]
  ??sl_gpu11-home 253:2    0 183.5G  0 lvm  /home


I just created

/home/scratch2

on /dev/sdb1

After the scheduled power outage, you should pull the dead drive out. I
don't recall the drive bay label (not a file server so not very important),
but since the server is going to be off, make sure you leave the NVMe drive
and a 3.5" HDD with the following serial number.



=== GOOD DRIVE INFORMATION SECTION ===
Model Family:     Seagate Enterprise Capacity 3.5 HDD
Device Model:     ST2000NM0055-1V4104
Serial Number:    ZC22SY3F

Cheers,
Predrag

On Thu, May 4, 2023 at 1:48?PM Piotr Bartosiewicz <pbartosi at andrew.cmu.edu>
wrote:

> Problem with the scratch directory.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230504/0e47a887/attachment.html>

From awd at cs.cmu.edu  Fri May  5 14:06:00 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Fri, 5 May 2023 14:06:00 -0400
Subject: Fwd: Reminder - Thesis Proposal - May 5, 2023 - Cristian Challu -
 Representing Time: Towards Pragmatic Multivariate Time Series Models
In-Reply-To: <b3d2f9c9-e932-67d8-0aad-6c0b523b74e2@andrew.cmu.edu>
References: <b3d2f9c9-e932-67d8-0aad-6c0b523b74e2@andrew.cmu.edu>
Message-ID: <CAJvAoytvZhWrqminNZ5o5GT-4fbnVehNEnn+bUZDSvWN0chrbw@mail.gmail.com>

Reminder: in less than an hour we will have the final culmination of a week
loaded with milestone talks by multiple Autonians.

Cristian will entertain us with his thesis proposal. He's got a load of
cool ideas to present that many of us will find relevant to their own work.

Also, this may be historically the first Auton Lab thesis proposal talk
delivered from Africa.

Cheers,
Artur

---------- Forwarded message ---------
From: Diane Stidle <stidle at andrew.cmu.edu>
Date: Fri, May 5, 2023 at 1:31?PM
Subject: Reminder - Thesis Proposal - May 5, 2023 - Cristian Challu -
Representing Time: Towards Pragmatic Multivariate Time Series Models
To: ml-seminar at cs.cmu.edu <ML-SEMINAR at cs.cmu.edu>, <ywu at stat.ucla.edu>


*Thesis Proposal*

Date: May 5, 2023
Time: 3:00pm (EST) (Remote only)

*Title: **Representing Time: Towards Pragmatic Multivariate Time Series
Models*
Abstract:
Time series models leverage temporal dependencies among observations and
interactions between multiple features in a data stream. During the last
decade, the unprecedented success of deep learning (DL), primarily in its
application to Computer Vision and Natural Language Processing, has slowly
but steadily permeated to time series analytics. From Recurrent Neural
Networks to Transformers, new advancements in architectural design improved
capabilities and performance. Despite this success, I identify several
challenges to the broader adoption of current state-of-the-art (SoTA) time
series models, including distribution shifts, missing data, computational
complexity, trustworthiness, and interpretability. The success of DL models
is usually justified by their ability to automatically discover helpful
data representations. We typically hope that multivariate time series
models, that involve high-dimensional objects with numerous time series and
temporal observations, exhibit useful temporal dependencies and
inter-feature relations. I propose to design new DL architectures and
algorithms for forecasting and anomaly detection tasks that leverage these
dependencies to induce efficient learning of representations capable of (i)
improving the models' performance, (ii) improving robustness by favoring
domain adaptation, and (iii) reducing overparameterization to improve
scalability. The already completed work includes five different models and
algorithms that achieve SoTA performance on relevant tasks and address a
few key adoption challenges. The proposed research extends that work in the
following directions: developing healthcare applications, analyzing the
enabling conditions for transfer learning between domains and tasks, and
extending prior work to multivariate long-horizon forecasting.

*Thesis Committee:*
Artur Dubrawski, Chair
Roni Rosenfeld
Barnabas Poczos
Ying Nian Wu (UCLA)

Link to the draft document:
https://drive.google.com/drive/u/3/folders/1fyyY_h9Ok3zk1a5KzTgsJNCq4mIRhh9G

Zoom meeting link:
https://cmu.zoom.us/j/92381510152?pwd=ZXk0Q0JaNlVVSk55a3RXZkM1c2ZYQT09

-- 
Diane Stidle
PhD Program Manager
Machine Learning Department
Carnegie Mellon Universitystidle at andrew.cmu.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230505/f7e4ac06/attachment.html>

From arundhat at andrew.cmu.edu  Mon May  8 17:13:16 2023
From: arundhat at andrew.cmu.edu (Arundhati Banerjee)
Date: Mon, 8 May 2023 17:13:16 -0400
Subject: GPU22 scratch is full
Message-ID: <7C3CCC4A-E3BA-4C67-9F5B-22AF38551682@andrew.cmu.edu>

Hi,

Since the scratch is full, could you please remove any files that you don?t need on gpu22? Thank you!

Best,
Arundhati

From awd at cs.cmu.edu  Tue May  9 09:50:07 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Tue, 9 May 2023 09:50:07 -0400
Subject: Fwd: SCS Town Hall for CMU's Open Science and Open Source Programs -
 June 8 at 1:00pm
In-Reply-To: <CAL+g+9CWt8U91naRddZcofzSpHO-nfqf_SXdU2-GY-CGH1csPA@mail.gmail.com>
References: <CAL+g+9CWt8U91naRddZcofzSpHO-nfqf_SXdU2-GY-CGH1csPA@mail.gmail.com>
Message-ID: <CAJvAoyshb-QauaSEhw6idBsW8r+BKVXETMNOihymubWAv6-+Fg@mail.gmail.com>

This may be of broad interest to the Lab, in particular to those of us who
already do or who are considering publishing their code and methods in the
open.

Cheers
Artur

---------- Forwarded message ---------
From: Nichole Merritt <ncmerritt at cmu.edu>
Date: Tue, May 9, 2023 at 9:09?AM
Subject: SCS Town Hall for CMU's Open Science and Open Source Programs -
June 8 at 1:00pm
To: <scs-faculty at cs.cmu.edu>


Dear SCS Faculty,

The CMU Libraries Open Science and Open Source Programs will host a virtual
town hall for SCS faculty on June 8 at 1:00pm. They will provide an
overview of the services and tools available to support areas of your
research and any class efforts, and answer any questions you may have.


*Thursday, June 8, 2023*
*1:00 ? 2:00 p.m. ET*
*Zoom Link: *
https://cmu.zoom.us/j/98236449319?pwd=SllXa1lmVFBDVEJWR3dlQ0xlN1Y4QT09
*[Add to Google Calendar
<https://calendar.google.com/calendar/event?action=TEMPLATE&tmeid=M2hwZzVoNDc2ZjVxNDh0a3AwMWwyN2FyaWkgbmljaG9sZW1AYW5kcmV3LmNtdS5lZHU&tmsrc=nicholem%40andrew.cmu.edu>]*



The Open Science <https://www.library.cmu.edu/services/open-science> team
will cover the following topics:

   - Introduce open science and Open Science Program
   - Introduce open-access publishing support
   - Introduce tools and services to enable data management and sharing
   - Introduce open-source resources which can help supplement classes and
   offer scaffolded training
   - Introduce data consultation services which, for example, can support
   less familiar research methods


The Open Source
<https://www.cmu.edu/news/stories/archives/2022/july/sloan-foundation-funds-new-office-to-support-open-source-activities>
team will cover the following topics:

   - Introduce open source programs office (OSPO)
   - Identify initial open source related topics of interest such as
   licensing, education, and memory safe programming
   - Mention initial outreach to the student population with the intent to
   seek out advice, feedback, etc.


-- 
Nichole Merritt
Assistant Director of Administration
School of Computer Science
Carnegie Mellon University
o: 412.268.1659  ?  m: 412.260.9682
_______________________________________________
ri-voting mailing list
ri-voting at lists.andrew.cmu.edu
https://lists.andrew.cmu.edu/mailman/listinfo/ri-voting
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230509/c293997c/attachment.html>

From awd at cs.cmu.edu  Tue May  9 21:01:27 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Tue, 9 May 2023 21:01:27 -0400
Subject: Fwd: MLD TA Awards
In-Reply-To: <de5a5ed0-958c-2edb-faa4-1b044993e894@andrew.cmu.edu>
References: <CAL0SxtzQwdQ3UGVWEgCnJG_Dfs9Mjo-2U9od6iebhDyWz0Abfw@mail.gmail.com>
 <de5a5ed0-958c-2edb-faa4-1b044993e894@andrew.cmu.edu>
Message-ID: <CAJvAoyvWuWYV1+yfNMDGpBO_G6qAfz65rDm4dS5-DA60U=MjsA@mail.gmail.com>

Another good day for the Auton Lab!

Two Autonians (one of them our alumna) out of 10 awardees in total were
recognized for their exceptional duty as Teaching Assistants for MLD.

Congrats Ian and Sana!

Artur

---------- Forwarded message ---------
On 5/8/2023 9:37 AM, Matt Gormley wrote:

Dear MLD TAs, students, faculty, and staff,

It is my great pleasure to announce the MLD TA Award recipients. We
presented awards covering the last two academic years at the MLD
End-of-year Picnic. Congratulations to all!

MLD TA of the Year (2022 - 2023)

   -

   Abhishek Vijayakumar - 10301/601 F22, S23


MLD TA Award (2022 - 2023)

   -

   Jennifer Hsia - 10708 S23
   -

   Nari Johnson - 10701 S23
   -

   Deep Patel - 10315 S23, F22
   -

   Olivier Filion - 10417/617 F22


MLD TA of the Year (2021 - 2022)

   -

   Ian Char - 10606/607 F21, 10716 S20


MLD TA Award (2021 - 2022)

   -

   Eric Liang - 10701 S22, F21, 10418 F22
   -

   Sana Lakdawala - 10301/601 S22, F21


Cheers,
-Matt
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230509/8752ff04/attachment.html>

From predragp at andrew.cmu.edu  Thu May 11 09:38:00 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Thu, 11 May 2023 09:38:00 -0400
Subject: Increasing shared memory limits on gpu25/27
In-Reply-To: <CADAXqzNdsOO0ziivpQGNyZDnWASxVcrxa=Qsqj7sz_7AqibuYQ@mail.gmail.com>
References: <CADAXqzNdsOO0ziivpQGNyZDnWASxVcrxa=Qsqj7sz_7AqibuYQ@mail.gmail.com>
Message-ID: <CAEFFt-FH0w-H3JPjGaG-UF_vqzf7TE4p5z9qU2Db+vyPkvkHvw@mail.gmail.com>

Hi Brian,

The honest answer is that I have no idea whether or not increasing shared
memory segments is a good or a bad idea. The good news is that one of our
staff members, Piggy Yarroll, has done some significant work on the Linux
kernel, so he might be able to give us some ideas.  I would not be
surprised to find out that some other Auton Lab account holders have a much
better idea what is the right cause of action than me.

My knee jerk reaction is that kernel variables are set there for a reason.
The prevailing philosophy in the OpenBSD camp to which I belong is that one
should not touch them unless one really knows what she/he is doing. I don't
:-( The only tunables I have ever played in the Auton Lab are the ZFS ZIL
and SLOG.

https://www.truenas.com/blog/zfs-zil-and-slog-demystified/

In my personal experience, it takes a real computer scientist (which I am
not) even to correctly measure performance. The most benchmarks I have seen
around the internet are just plain wrong and the tests are set up
incorrectly.  If I really knew anything about squeezing performance out of
the system I would be making big bucks working in industry :-)

Best,
Predrag



On Thu, May 11, 2023 at 3:35?AM Brian Yang <brianyan at andrew.cmu.edu> wrote:

> Hi Predrag,
>
> I've been running into issues with loading larger data batches on these
> machines to make use of the bigger GPUs. Basically we seem to have ample
> RAM / swap / shared memory / GPU memory available to load much larger
> batches, but I think I'm running up against the system-wide limit on the
> number of shared memory segments.
>
> Would it be possible for us to increase the number of shared memory
> segments allowed on GPU25 / 27, maybe from 4096 to 16384?
>
> I checked the current limit with "cat /proc/sys/kernel/shmmni". I think
> you can change the limit with "echo 16384 > /proc/sys/kernel/shmmni",
> although you would also need to do something like "echo
> "kernel.shmmni=16384" >> /etc/sysctl.conf" to make the change persist on
> rebooting.
>
> If there's something I'm missing / a reason why increasing the limit is a
> bad idea, please let me know. There might be a way to rewrite my data
> loader to avoid this issue, but I think this would be much easier if it's
> doable without major side effects.
>
> Thanks,
> Brian
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230511/62124034/attachment.html>

From predragp at andrew.cmu.edu  Fri May 12 13:54:13 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Fri, 12 May 2023 13:54:13 -0400
Subject: is there outage?
In-Reply-To: <CABwU0nmYuPJos0qYr8MRcz2z0kDn5bjPkm35k0YrkkugAYvdsQ@mail.gmail.com>
References: <CABwU0nmYuPJos0qYr8MRcz2z0kDn5bjPkm35k0YrkkugAYvdsQ@mail.gmail.com>
Message-ID: <CAEFFt-GDeVTRGcz6Ai12b7FR_7Hnj7fz0X2BA5h=FHM5-iKQQw@mail.gmail.com>

All ZFS pools look ok. All servers look OK. I will log into every
individual server manually and check the status of the daemons. I suspect
the machines got overloaded and the autofs daemon got killed. I will give
you an update shortly.

Predrag

On Fri, May 12, 2023 at 11:44?AM Ifigeneia Apostolopoulou <
iapostol at andrew.cmu.edu> wrote:

> many gpus give directory /zfsauton3/home/iapostol: No such file or
> directory!
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230512/b944d573/attachment.html>

From predragp at andrew.cmu.edu  Fri May 12 15:33:48 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Fri, 12 May 2023 15:33:48 -0400
Subject: A Lot of Clusters are Non-functional
In-Reply-To: <CAEm4dA6LrKuPG_E4hT3hMCXaNJRz6sgiu+N-X8QKSb2V8RzAqg@mail.gmail.com>
References: <CAEm4dA6LrKuPG_E4hT3hMCXaNJRz6sgiu+N-X8QKSb2V8RzAqg@mail.gmail.com>
Message-ID: <CAEFFt-GDfQOgms4bLbm0iMDNe0CV0SksOLk_D6HMxxxW4zkcFw@mail.gmail.com>

Hi Chenghui,


Let me clarify my previous post. The number of machines were zombified by
users, not by me or some other supernatural power. I know better than to
change any configuration a few days before the major conference.

As you might know, the cluster is a shared resource. I have zero control
over the behaviour of such a large, diverse group. I can  advise,
recommend, or educate about the topics I know something about. I don't
debug other people's code. I don't have any magic powers.

Predrag

On Fri, May 12, 2023 at 3:24?PM Chenghui Zhou <chenghuz at andrew.cmu.edu>
wrote:

> Hello Predrag,
>
> As NeurIPS is nearing, it is really important for us to have clusters
> functioning to run some of the lat experiments. However, right now, I
> notice that at least 5 clusters that I'm able to log in do not have my
> working directory in it, same issue as Ifi reported. The rest of the
> functional clusters are all full. Could you please look into that, thank
> you!
>
> Chenghui
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230512/6d082e14/attachment.html>

From chenghuz at andrew.cmu.edu  Fri May 12 15:35:45 2023
From: chenghuz at andrew.cmu.edu (Chenghui Zhou)
Date: Fri, 12 May 2023 15:35:45 -0400
Subject: A Lot of Clusters are Non-functional
In-Reply-To: <CAEFFt-GDfQOgms4bLbm0iMDNe0CV0SksOLk_D6HMxxxW4zkcFw@mail.gmail.com>
References: <CAEm4dA6LrKuPG_E4hT3hMCXaNJRz6sgiu+N-X8QKSb2V8RzAqg@mail.gmail.com>
 <CAEFFt-GDfQOgms4bLbm0iMDNe0CV0SksOLk_D6HMxxxW4zkcFw@mail.gmail.com>
Message-ID: <CAEm4dA6OsCTdYBEP4VhGHZe-KxHBuu4aHZehw0FQf+Sbs4d9tQ@mail.gmail.com>

Thanks for looking into it! They seem to be working now!

On Fri, May 12, 2023 at 3:34?PM Predrag Punosevac <predragp at andrew.cmu.edu>
wrote:

> Hi Chenghui,
>
>
> Let me clarify my previous post. The number of machines were zombified by
> users, not by me or some other supernatural power. I know better than to
> change any configuration a few days before the major conference.
>
> As you might know, the cluster is a shared resource. I have zero control
> over the behaviour of such a large, diverse group. I can  advise,
> recommend, or educate about the topics I know something about. I don't
> debug other people's code. I don't have any magic powers.
>
> Predrag
>
> On Fri, May 12, 2023 at 3:24?PM Chenghui Zhou <chenghuz at andrew.cmu.edu>
> wrote:
>
>> Hello Predrag,
>>
>> As NeurIPS is nearing, it is really important for us to have clusters
>> functioning to run some of the lat experiments. However, right now, I
>> notice that at least 5 clusters that I'm able to log in do not have my
>> working directory in it, same issue as Ifi reported. The rest of the
>> functional clusters are all full. Could you please look into that, thank
>> you!
>>
>> Chenghui
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230512/c1e31d07/attachment.html>

From predragp at andrew.cmu.edu  Fri May 12 15:41:33 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Fri, 12 May 2023 15:41:33 -0400
Subject: A Lot of Clusters are Non-functional
In-Reply-To: <CAEm4dA6OsCTdYBEP4VhGHZe-KxHBuu4aHZehw0FQf+Sbs4d9tQ@mail.gmail.com>
References: <CAEm4dA6LrKuPG_E4hT3hMCXaNJRz6sgiu+N-X8QKSb2V8RzAqg@mail.gmail.com>
 <CAEFFt-GDfQOgms4bLbm0iMDNe0CV0SksOLk_D6HMxxxW4zkcFw@mail.gmail.com>
 <CAEm4dA6OsCTdYBEP4VhGHZe-KxHBuu4aHZehw0FQf+Sbs4d9tQ@mail.gmail.com>
Message-ID: <CAEFFt-GPp05fHB1yMc2yJByuK93S9vszByOFSaObnBOZfctMsQ@mail.gmail.com>

There is nothing to be thankful about. A few days worth of work was lost
due to misuse, lack of debugging, and poor design of experiments. Just
imagine the loss of revenue if these were iTunes servers running holiday
sales...

Predrag

On Fri, May 12, 2023 at 3:36?PM Chenghui Zhou <chenghuz at andrew.cmu.edu>
wrote:

> Thanks for looking into it! They seem to be working now!
>
> On Fri, May 12, 2023 at 3:34?PM Predrag Punosevac <predragp at andrew.cmu.edu>
> wrote:
>
>> Hi Chenghui,
>>
>>
>> Let me clarify my previous post. The number of machines were zombified by
>> users, not by me or some other supernatural power. I know better than to
>> change any configuration a few days before the major conference.
>>
>> As you might know, the cluster is a shared resource. I have zero control
>> over the behaviour of such a large, diverse group. I can  advise,
>> recommend, or educate about the topics I know something about. I don't
>> debug other people's code. I don't have any magic powers.
>>
>> Predrag
>>
>> On Fri, May 12, 2023 at 3:24?PM Chenghui Zhou <chenghuz at andrew.cmu.edu>
>> wrote:
>>
>>> Hello Predrag,
>>>
>>> As NeurIPS is nearing, it is really important for us to have clusters
>>> functioning to run some of the lat experiments. However, right now, I
>>> notice that at least 5 clusters that I'm able to log in do not have my
>>> working directory in it, same issue as Ifi reported. The rest of the
>>> functional clusters are all full. Could you please look into that, thank
>>> you!
>>>
>>> Chenghui
>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230512/ed16d269/attachment-0001.html>

From predragp at andrew.cmu.edu  Tue May 16 10:58:54 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Tue, 16 May 2023 08:58:54 -0600
Subject: SCS Computing: An extended power outage will affect your
 machine-room hosted systems on 5/20/2023.
In-Reply-To: <CAEFFt-Gs0YDTx2b8E0uSvxk=1qbx-UcAnqqGuf4Z=KtjBpcE=g@mail.gmail.com>
References: <64525c6980ef5_1ad742aad9f65e5c428375@massmail-02.andrew.cmu.edu.mail>
 <CAEFFt-Gs0YDTx2b8E0uSvxk=1qbx-UcAnqqGuf4Z=KtjBpcE=g@mail.gmail.com>
Message-ID: <CAEFFt-Eq9juFOjKZzVYRKduRknc9Rn5aAVbJFPG452fLZ2iw0A@mail.gmail.com>

Just a gentle reminder that the lab cluster will be powered of on Friday
afternoon due to the scheduled power outage.

Predrag

On Wed, May 3, 2023, 7:29 AM Predrag Punosevac <predragp at andrew.cmu.edu>
wrote:

> Please book your calendar. I will not be physically present in Pittsburgh.
> You need to go and power up critical infrastructure machines.
>
> Predrag
>
> ---------- Forwarded message ---------
> From: <help at cs.cmu.edu>
> Date: Wed, May 3, 2023, 9:06 AM
> Subject: SCS Computing: An extended power outage will affect your
> machine-room hosted systems on 5/20/2023.
> To: <predragp at cs.cmu.edu>
>
>
> *FMS is planning an extended power outage and we will shut down your
> systems in our Wean Hall machine room for the duration.*
>
> You are a contact on systems that will be shutdown when FMS suspends
> electrical service to multiple campus buildings, including Wean Hall as the
> new Highmark Center for Health, Wellness and Athletics is connected to the
> campus power grid.
>
> *We will begin taking systems offline at 4:00am on Saturday, May 20, 2023.*
> Once power restoration is confirmed, we plan to power-on systems after
> 12:00pm (noon) on Sunday, May 21, 2023. Some systems may be powered down
> longer for maintenance.
>
> To verify the authenticity of this message, visit our website
> <https://computing.cs.cmu.edu/dashboard/2023/scs-power-outage-alert-5-20-2023>
> .
>
> If you have any questions, reply to this message, contact the SCS Help
> Desk at 412-268-4231 or send mail to help at cs.cmu.edu.
>
>
>
> *SCS Computing Facilities*
>
> https://computing.cs.cmu.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230516/0bb75adb/attachment.html>

From awd at cs.cmu.edu  Tue May 16 23:10:03 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Tue, 16 May 2023 23:10:03 -0400
Subject: Dr. Kin Gutierrez Olivares is staying with the Auton Lab as a Project
 Scientist!
Message-ID: <CAJvAoyvt4Ph33jA_VSL5292-R8uvifozU1hJ5c40_zfqDO353g@mail.gmail.com>

Team,

I am very pleased to let you all know that Kin is staying with us as a
member of the Auton Lab research staff, initially for about another year.
This is great news.
I am looking forward to his continued dominance in the area of time series
forecasting on the research side, and to leveraging his expertise in
accelerating many of our applied projects that can greatly benefit from his
involvement.

Please join me in welcoming one of our newest doctors to our team in his
new role!

Kin's office will be NSH 3124, he will be sharing it with Angela Chen and
with one of our incoming new hires coming onboard soon.

Cheers,
Artur
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230516/8b8354c7/attachment.html>

From awd at cs.cmu.edu  Wed May 17 00:06:21 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Wed, 17 May 2023 00:06:21 -0400
Subject: Congrats to our own Professor Karen Chen and her team at UMBC!
Message-ID: <CAJvAoysROyNG9CN3ASd+s6R5+=9dat4VqL-TWugoCO+u+PwQrg@mail.gmail.com>

Dear Autonians,

Check this out:
https://tools-competition.org/announcing-finalists-for-darpas-ai-tools-for-adult-learning/

Karen has continued developing further the "caselet" toolkit which we
originally developed for use in my data science course teaching to enhance
hands-on skills of aspiring data scientists. It gained a somewhat mixed
reputation among CMU graduate students who were exposed to it: tough and
demanding, but helpful. I hear that Karen has kept it at the high level of
ambitiousness while improving and expanding its scope, after she moved on
after graduating with the Auton Lab PhD to become faculty at the University
of Maryland Baltimore County.

Now, the toolkit was selected as a finalist in a tough DARPA competition,
out of 250 entries. Way to go Karen! Big congrats to you and your team.
We will be cheering for you and the caselets moving forward.

Cheers,
Artur
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230517/746af114/attachment.html>

From awd at cs.cmu.edu  Thu May 18 20:41:48 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Thu, 18 May 2023 20:41:48 -0400
Subject: This seminar series may be of interest to many Autonians
Message-ID: <CAJvAoys+ukwaaGz4XPhDhopTrx6HkNg-Nz7vf1w+0ZBqohYDcQ@mail.gmail.com>

https://www.cs.cmu.edu/generative-ai/events
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230518/aaf21747/attachment.html>

From arundhat at andrew.cmu.edu  Fri May 19 13:31:12 2023
From: arundhat at andrew.cmu.edu (Arundhati Banerjee)
Date: Fri, 19 May 2023 13:31:12 -0400
Subject: GPU25 scratch is full
Message-ID: <09CFDD12-F990-449D-90BA-E9D8E1C36109@andrew.cmu.edu>

Hi,

Since the scratch is full, could you please remove any files that you don?t need on gpu25? Thank you!


-Arundhati

From predragp at andrew.cmu.edu  Sun May 21 19:19:36 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Sun, 21 May 2023 19:19:36 -0400
Subject: Cluster is back online, fully functional
Message-ID: <CAEFFt-FBNe1bHSdzM8zqYGwFNd2wm827+_NhPh-kNAZi18zpPQ@mail.gmail.com>

Dear Autonians,

The cluster is back on-line and fully functional. The majority of work was
done by Piotr with a bit of my assistance (ssh and the phone). Currently
only two computing nodes are not operational.

lov3 needs a new power supply. It will get it tomorrow morning as Piotr has
to eat and sleep after 8h in the server room.

gpu1 most likely has a dead RAM module. It is not booting. Piotr will see
if we have a spare one or if we have to get it on Ebay (the machine is
almost 10 years old).  I am not concerned about it. We will get more life
out of that machine.

Best,
Predrag
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230521/c5a1c1e3/attachment.html>

From awd at cs.cmu.edu  Sun May 21 20:04:33 2023
From: awd at cs.cmu.edu (Artur Dubrawski)
Date: Sun, 21 May 2023 20:04:33 -0400
Subject: Cluster is back online, fully functional
In-Reply-To: <CAEFFt-FBNe1bHSdzM8zqYGwFNd2wm827+_NhPh-kNAZi18zpPQ@mail.gmail.com>
References: <CAEFFt-FBNe1bHSdzM8zqYGwFNd2wm827+_NhPh-kNAZi18zpPQ@mail.gmail.com>
Message-ID: <CAJvAoyssNdawQWbeQ5CJ8kUo97ZNMJ-kat7UD_3YXHuja-pmXA@mail.gmail.com>

This is amazing, whenever anyone shuts down a system of such complexity,
many things can go wrong when the components are being brought back up.
Thank you very much Predrag and Piotr and congratulations on the job well
done!

Cheers
Artur


On Sun, May 21, 2023 at 7:19?PM Predrag Punosevac <predragp at andrew.cmu.edu>
wrote:

> Dear Autonians,
>
> The cluster is back on-line and fully functional. The majority of work was
> done by Piotr with a bit of my assistance (ssh and the phone). Currently
> only two computing nodes are not operational.
>
> lov3 needs a new power supply. It will get it tomorrow morning as Piotr
> has to eat and sleep after 8h in the server room.
>
> gpu1 most likely has a dead RAM module. It is not booting. Piotr will see
> if we have a spare one or if we have to get it on Ebay (the machine is
> almost 10 years old).  I am not concerned about it. We will get more life
> out of that machine.
>
> Best,
> Predrag
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230521/fe5131c7/attachment.html>

From kelenber at andrew.cmu.edu  Sun May 21 21:26:27 2023
From: kelenber at andrew.cmu.edu (Kimberly Elenberg)
Date: Sun, 21 May 2023 21:26:27 -0400
Subject: Congrats to our own Professor Karen Chen and her team at UMBC!
In-Reply-To: <CAJvAoysROyNG9CN3ASd+s6R5+=9dat4VqL-TWugoCO+u+PwQrg@mail.gmail.com>
References: <CAJvAoysROyNG9CN3ASd+s6R5+=9dat4VqL-TWugoCO+u+PwQrg@mail.gmail.com>
Message-ID: <CAHycrtP1n+Mqr4pTObUsxckiX=aozWv_yNB3NCvS_4est2km3w@mail.gmail.com>

Awesome Karen!!! Congratulations!

On Wed, May 17, 2023 at 12:08?AM Artur Dubrawski <awd at cs.cmu.edu> wrote:

> Dear Autonians,
>
> Check this out:
>
> https://tools-competition.org/announcing-finalists-for-darpas-ai-tools-for-adult-learning/
>
> Karen has continued developing further the "caselet" toolkit which we
> originally developed for use in my data science course teaching to enhance
> hands-on skills of aspiring data scientists. It gained a somewhat mixed
> reputation among CMU graduate students who were exposed to it: tough and
> demanding, but helpful. I hear that Karen has kept it at the high level of
> ambitiousness while improving and expanding its scope, after she moved on
> after graduating with the Auton Lab PhD to become faculty at the University
> of Maryland Baltimore County.
>
> Now, the toolkit was selected as a finalist in a tough DARPA competition,
> out of 250 entries. Way to go Karen! Big congrats to you and your team.
> We will be cheering for you and the caselets moving forward.
>
> Cheers,
> Artur
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230521/6f955c26/attachment.html>

From predragp at andrew.cmu.edu  Tue May 23 14:02:22 2023
From: predragp at andrew.cmu.edu (Predrag Punosevac)
Date: Tue, 23 May 2023 12:02:22 -0600
Subject: bash down
Message-ID: <CAEFFt-ERXOpR-sOfkueFp76xXSxF5vZH78mT+9sv3pc2HD1psw@mail.gmail.com>

HDD died. Piotr is fixing it.

Predrag
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/autonlab-users/attachments/20230523/db7d312b/attachment.html>