Categories and tags knowledge base for local files

 I had planned to write utility to store documents metadata, stumble upon an kb, a text-oriented minimalist command line knowledge base manager. Let's imagine you have a lot of text and graphics in your system, you need to classify to quick search on it. These documents are known as artifacts, because they all add hyper map to your knowledge then you can obtain it in time, because you know a category or even tag. In my experience, work knowledge base contain thousands of documents and searching with this type of invisible markers make data extraction possible and in many cases incredible fast. So that, how to install KB in Oracle linux:

curl "https://bootstrap.pypa.io/get-pip.py" -o "get-pip.py"
python3 get-pip.py
pip install kb-manager

On ubuntu 20.04

sudo apt install python3-pip

pip3 install kb-manager

Then, kb is located in ~/.local/bin

.NET Core debug, and UML diagrams on linux Visual Code

There is an important to debug .Net Core programs under linux. I thought it's should be different setup for linux OS, but no, thank you for Pluralsight article one sequence for every OS:

1. Add .NET Core Extension Pack.

2. Press Ctrl-Shift-P to open shortcuts to find .NET: Generate Assets for Build and Debug

3. Then use debug pane

I hope debug will help to finish program writing  quicker and then you want to writer documentation using UML diagrams. As for me it is hard to draw UML with vector primitives, it is better to utilize PlantUML

 Extensions to auto generate  uml class diagrams from project code:

1. CSharp to PlantUML - this help to generate all classes in separate .puml files, then you can copy it to your diagram instead of write from scratch

 

2. PlantUML - this extension helps with intelliSense and preview.

You need to install graphviz in your system too.

Its looks like we have ere of Visual Code for linux workspaces for most of languages!

Utils in Graphviz package

 There is a list of utils in graphviz:

acyclic - make directed graph acyclic
bcomps - biconnected components filter for graphs
ccomps - connected components filter for graphs
circo - filter for circular layout of graphs
cluster - find clusters in a graph and augment the graph with this information.
diffimg - Calculates intersection between two images
dijkstra - single-source distance filter
dot - filter for drawing directed graphs
gxl2gv,gv2gxl - GXL-GV converters
dotty - A Customizable Graph Editor
edgepaint - edge coloring to disambiguate crossing edges
fdp - filter for drawing undirected graphs
gc - count graph components
gml2gv,gv2gml - GML-DOT converters
graphml2gv - GRAPHML-DOT converter
gvcolor - flow colors through a ranked digraph
gvgen - generate graphs
gvmap - find clusters and create a geographical map highlighting clusters.
gvpack - merge and pack disjoint graphs
gvpr - graph pattern scanning and processing language
lefty - A Programmable Graphics Editor
lneato - A Customizable Graph Editor
mm2gv - Matrix Market-DOT converters
neato - filter for drawing undirected graphs
nop - pretty-print graph file
osage - filter for drawing clustered graphs
patchwork - filter for drawing clustered graphs as treemaps
prune - Prune directed graphs
sccmap - extract strongly connected components of directed graphs
sfdp - filter for drawing large undirected graphs
tred - transitive reduction filter for directed graphs
twopi - filter for radial layouts of graphs
unflatten - adjust directed graphs to improve layout aspect ratio
vimdot - Combined text editor and dot viewer 

I retrieved it with yum util repoquery:

yum -y install yum-utils
repoquery --list graphviz

Of course, I started with dot for me, but pretty look for all available capabilities in grapviz too.

Simple scenario is a image from dot file:

digraph {
    a -> b;
    b -> c;
    c -> d;
    d -> a;
}

I took this graph from examples. It is easy to generate such graph in C# code today. Then I want to convert it to image, I found this  command:

dot ./1.dot -Tpng > ./1.png

Grayscale view for Gnome in Oracle linux

 I think, a lot of us want to concentrate on their pc work and especially for me colors are distraction. It is easy do disable color filters on Windows, but it is also possible to disable it and make display graysacle in linux too. There are commands to add 'Desaturate all' to gnome shell.

yum install gnome-tweak-tool

git clone https://gitlab.com/JanMa/desaturate_all.git

mv ~/Downloads/desaturate_all "/home/eugene/.local/share/gnome-shell/extensions"

Alt + F2 then r to restart DE

Enable Desaturate all extension in Tweaks app.

Use small 'three circles' icon on the right top corner to enable grayscale.

What I stadied with one year O’Reilly Learning

 It is clear for me, all publishers take their tiers in education process. And O’Reilly is not a beginner in understanding of computer science subject. So that, I can recommend to use publishers as the stairs for unknown technologies or theory and final stair is  O’Reilly. In a year I read 20 books, participate in 2 online training and the learning path. The most important thing for me was to realize that there are people who think like me and these people are experts. I can learn logic and pragmatism from them.

This is my recap from Learning Path: Graph Algorithms, By Mark Needham

1. Using a special framework, such as spark, allows you to focus on selecting a method to solve the problem and quickly determine the correctness of the idea, instead of guessing the only correct approach.

2. The importance of linked data is difficult to override, because the collected relationships allow further use of very simple decision-making algorithms. This means that relationships are a step towards simplifying decisions.

3. Determined algorithms are characteristic of generalization, and they cannot always find solutions for small groups in a variety of settings, but they allow you to see the big picture. Probabilistic solutions can improve the overall picture.

4. The graphs above the graphs and the super category are the key to determine the initial solution path, and these structures can be distinguished from the data.

5. It is the iterative approach in determining the desired algorithm, rather than a single selected approach, that allows you to achieve a solution that will reliably solve the problem, thus experimenting and making conclusions, this is the key to creating reliable solutions, instead of a completely theoretical approach.

Thank you, O’Reilly Learning for this year with me. 

 

   

Setup Visual Graph Editor in Oracle linux

 Thank you for simplified installing process with snapcraft I can install QVGE for my new Checklists project, I need to architect.

It might be expensive to setup yet another package manager to install one app as same as install second game to a system. I copy scenario from snapcraft here:

sudo dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm sudo dnf upgrade

 

sudo rpm -ivh https://dl.fedoraproject.org/pub/epel/epel-release-latest-7.noarch.rpm
 

sudo yum install snapd

 

sudo systemctl enable --now snapd.socket

 

sudo ln -s /var/lib/snapd/snap /snap

 

sudo snap install qvge

 

Where to find QVGE then:




/snap/bin/qvge  







This graph editor is very promising for every day modeling. I got memory errors with graphviz conversion and opening, I think, I need to wait new version, but nearer graph editor gephi is more scientific and not adopted to manual work with graphs. 







I also should recommend yEed Graph editor, it can be used for prototype product features through multiple components. It is really easy to draw feature vision directly on the meeting.


One I know about these great graph editors, they are producing very complicated files for diagrams. To integrate editor in program flow, I found TikZit is winner. Take a look on this file:

\begin{tikzpicture}
    \begin{pgfonlayer}{nodelayer}
        \node [style=none] (1) at (1.5, 4.5) {};
        \node [style=none] (2) at (0.75, 0) {};
        \node [style=none] (3) at (-10.5, 5) {dd};
        \node [style=none] (4) at (1.5, 4.5) {ss};
        \node [style=none] (5) at (3.5, -3) {};
        \node [style=none] (6) at (0.75, 0) {asdf};
        \node [style=none] (7) at (-6.25, -0.5) {asdf};
        \node [style=none] (8) at (-0.75, -2.25) {};
        \node [style=new style 0] (9) at (-7, 6.5) {};
        \node [style=new style 0] (10) at (-3.25, 5.5) {};
        \node [style=new style 0] (11) at (5, 8.25) {};
        \node [style=new style 0] (12) at (9.5, -1.5) {};
    \end{pgfonlayer}
    \begin{pgfonlayer}{edgelayer}
        \draw [bend right] (3.center) to (7.center);
        \draw [bend right] (4.center) to (7.center);
        \draw (4.center) to (6.center);
        \draw (6.center) to (8.center);
        \draw (6.center) to (5.center);
        \draw [style=new edge style 0, bend right=315] (8.center) to (7.center);
        \draw [style=new edge style 0, bend right=75, looseness=4.75] (9) to (10);
        \draw [style=new edge style 0, bend right] (10) to (11);
        \draw [style=new edge style 0, bend right=45, looseness=1.25] (11) to (12);
    \end{pgfonlayer}
\end{tikzpicture}

This elegant markup can produce two vertices with curved edge.

Ultracheap Pascal gpu on Oracle linux

It is time to setup working environment with Zotac P106-90 GPU as computation unit. I have installed latest CUDA 10.2 toolkit and driver to Orcale Linux without complicated tricks.

1. I download local installer CUDA run file from Nvidia.
2. I have changed my runlevel to multi-user (3) with

systemctl set-default multi-user.target

3. I rebooted to terminal only mutli-user runlevel

4. Trying to run ./cuda_10.2.89_440.33.01_linux.run few times, but in /var/log/nvidia-install.log I have seen X server still run.

5. Then I found in NVIDIA instruction I can kill X Server and remove lock file.

kill <process_id>
rm /tmp/.X2.lock

6. After resolve xserver lock I have a message in nvidia-installer.log that Nouveau driver is disabled and I need to reboot to run installation again to replace driver.

7. I repeated the trick with killing XServer process and lock file remove, I successfully install all Nvidia driver, CUDA Toolkit 10.2 and samples.

8. I changed  runlevel back to grapical

systemctl set-default graphical.level

9. I rebooted to DE to check cuda samples works.

Render Buddha fractal on decommissioned P106-90

Probably I got an answer, how to prepare me to parallel data computing with great Pascal architecture for my small budget 30usd. I have found Zotac P106-90 is out of mining game for now, I can bought and adopt for GPU computation. This graphics card looks like Zotac GTX 1060 but it have no any video output. My first test is for Windows 10, latest Cuda 10.2 produce great Buddha fractal for me. For render Buddha 48MB fractal image it takes 6 min.

nvcc --version

Cuda compilation tools, release 10.2, V10.2.89Built on Wed_Oct_23_19:32:27_Pacific_Daylight_Time_2019Copyright (c) 2005-2019 NVIDIA Corporationnvcc: NVIDIA (R) Cuda compiler driver

I used JCudaFractals written in Java with JCuda library.

mvn compile
mvn exec:java -Dexec.mainClass="net.marvk.jcudafractals.Main"

Sometime I got CUDA_ERROR_ILLEGAL_ADDRESS errors.

jcuda.CudaException: CUDA_ERROR_ILLEGAL_ADDRESS
    at jcuda.driver.JCudaDriver.checkResult (JCudaDriver.java:359)
    at jcuda.driver.JCudaDriver.cuCtxSynchronize (JCudaDriver.java:2139)
    at net.marvk.jcudafractals.fractal.Buddhabrot2.buddhabrot (Buddhabrot2.java:95)
    at net.marvk.jcudafractals.controller.Controller.lambda$new$0 (Controller.java:25)
    at java.util.stream.IntPipeline$1$1.accept (IntPipeline.java:180)
    at java.util.stream.Streams$RangeIntSpliterator.forEachRemaining (Streams.java:104)
    at java.util.Spliterator$OfInt.forEachRemaining (Spliterator.java:699)
    at java.util.stream.AbstractPipeline.copyInto (AbstractPipeline.java:484)
    at java.util.stream.AbstractPipeline.wrapAndCopyInto (AbstractPipeline.java:474)
    at java.util.stream.AbstractPipeline.evaluate (AbstractPipeline.java:550)
    at java.util.stream.AbstractPipeline.evaluateToArrayNode (AbstractPipeline.java:260)
    at java.util.stream.ReferencePipeline.toArray (ReferencePipeline.java:517)
    at net.marvk.jcudafractals.controller.Controller.<init> (Controller.java:26)
    at net.marvk.jcudafractals.Main.main (Main.java:10)
    at jdk.internal.reflect.NativeMethodAccessorImpl.invoke0 (Native Method)
    at jdk.internal.reflect.NativeMethodAccessorImpl.invoke (NativeMethodAccessorImpl.java:62)
    at jdk.internal.reflect.DelegatingMethodAccessorImpl.invoke (DelegatingMethodAccessorImpl.java:43)
    at java.lang.reflect.Method.invoke (Method.java:566)
    at org.codehaus.mojo.exec.ExecJavaMojo$1.run (ExecJavaMojo.java:282)
    at java.lang.Thread.run (Thread.java:834)

In common this cheap GPU works in calculations as expected.

C:\Users\Eugene>"C:\Program Files\NVIDIA Corporation\NVSMI\nvidia-smi.exe"
Thu Apr 30 22:49:50 2020
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 441.22       Driver Version: 441.22       CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name            TCC/WDDM | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  P106-090            TCC  | 00000000:02:00.0 Off |                  N/A |
| 47%   49C    P0    55W /  75W |    674MiB /  3012MiB |    100%      Default |
+-------------------------------+----------------------+----------------------+
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    0      1524      C   ...v7\win\64bit\Core_22.fah\FahCore_22.exe   289MiB |
|    0      2892      C   D:\CUDA-Z-0.10.251-32bit.exe                  73MiB |
|    0      3360      C   ...gram Files\Java\jdk-11.0.7\bin\java.exe   270MiB |
+-----------------------------------------------------------------------------+

I also check performance parameters in CUDA-Z program that show, it is slow to upload data to device, but it very fast to share data between several devices.

CUDA-Z Report
=============
Version: 0.10.251 32 bit http://cuda-z.sf.net/
OS Version: Windows AMD64 6.2.9200
Driver Version: 441.22 (TCC)
Driver Dll Version: 10.20 (26.21.14.4122)
Runtime Dll Version: 6.50
Core Information
----------------
 Name: P106-090
 Compute Capability: 6.1
 Clock Rate: 1531 MHz
 PCI Location: 0:2:0
 Multiprocessors: 5
 Threads Per Multiproc.: 2048
 Warp Size: 32
 Regs Per Block: 65536
 Threads Per Block: 1024
 Threads Dimensions: 1024 x 1024 x 64
 Grid Dimensions: 2147483647 x 65535 x 65535
 Watchdog Enabled: No
 Integrated GPU: No
 Concurrent Kernels: Yes
 Compute Mode: Default
 Stream Priorities: Yes
Memory Information
------------------
 Total Global: 3012.12 MiB
 Bus Width: 192 bits
 Clock Rate: 4004 MHz
 Error Correction: No
 L2 Cache Size: 48 KiB
 Shared Per Block: 48 KiB
 Pitch: 2048 MiB
 Total Constant: 64 KiB
 Texture Alignment: 512 B
 Texture 1D Size: 131072
 Texture 2D Size: 131072 x 65536
 Texture 3D Size: 16384 x 16384 x 16384
 GPU Overlap: Yes
 Map Host Memory: Yes
 Unified Addressing: No
 Async Engine: Yes, Bidirectional
Performance Information
-----------------------
Memory Copy
 Host Pinned to Device: 738.929 MiB/s
 Host Pageable to Device: 712.643 MiB/s
 Device to Host Pinned: 776.933 MiB/s
 Device to Host Pageable: 724.519 MiB/s
 Device to Device: 35.1506 GiB/s
GPU Core Performance
 Single-precision Float: 2149.6 Gflop/s
 Double-precision Float: 36.9897 Gflop/s
 64-bit Integer: 92.5176 Giop/s
 32-bit Integer: 743.671 Giop/s
 24-bit Integer: 558.109 Giop/s
Generated: Thu Apr 30 23:18:43 2020

Two AppImage for a day

I have found large Linux applications have packed for AppImages I can run without install.

The AppImage format is a format for packaging applications in a way that allows them to run on a variety of different target systems (base operating systems, distributions) without further modification.

 First of them is Gravit Designer that can open and edi Adobe Illustrator files, looks like important thing.

Second, is decade known for me program, Scilab. It works now pretty fast when I know it teen years ago.

To download AppImage for Scilab, base on the answer:

mkdir ~/Software
cd ~/Software
wget https://github.com/davidcl/Scilab.AppDir/releases/download/6.0.2-1/Scilab-x86_64.AppImage
chmod +x Scilab-x86_64.AppImage
ln -s Scilab-x86_64.AppImage scilab

Where to find space to Steam games on small user home

Usually, Linux user have limited space on their home for data, but not for large binaries what shipped with Steam. I have found an option to change steamapps folder to separate large volume belong to the user.

Steam library folders allow to setup location for games and finally I can enjoy new Life is Strange episode for me.

Install OpenCl Neo on Oracle Linux

I found it is easy to install opencl to Oracle Linux as one, two, three.

curl http://registrationcenter-download.intel.com/akdlm/irc_nas/11396/SRB5.0_linux64.zip > SRB5.0_linux64.zip
unzip SRB5.0_linux64.zip
yum install intel-opencl-cpu-r5.0-63503.x86_64.rpm    
yum install intel-opencl-devel-r5.0-63503.x86_64.rpm      
yum install intel-opencl-r5.0-63503.x86_64.rpm              

yum provides "*/libOpenCL.so"











Produced output with path to the lib:

intel-opencl-r5.0-63503.x86_64 : OpenCL ICD loader and ICD for Intel Iris Graphics
Repo        : @System
Matched from:
Filename    : /opt/intel/opencl/libOpenCL.so

Then test opencl runtime directly in .Net Core

git clone https://github.com/lecode-official/opencl-dotnet.git

cd opencl-dotnet

cd OpenCl.DotNetCore.Tests

dotnet run 

Supported Platforms & Devices:
+--------------------------------------------------+----------------+----------------------+-----------------------------------------+----------------+----------------+----------------+----------------
| Platform                                         | OpenCL Version | Vendor               | Device                                  | Driver Version | Bits           | Memory         | Clock Speed    | Available      |
+--------------------------------------------------+----------------+----------------------+-----------------------------------------+----------------+----------------+----------------+----------------
| Intel(R) OpenCL                                  | 2.0            | Intel(R) Corporation | Intel(R) Core(TM) i3-8100 CPU @ 3.60GHz | 1.2.0.475      | 64 Bit         | 3.43 GiB       | 3600 MHz       | ✔              |
+--------------------------------------------------+----------------+----------------------+-----------------------------------------+----------------+----------------+----------------+----------------
| Intel(R) CPU Runtime for OpenCL(TM) Applications | 2.1            | Intel(R) Corporation | Intel(R) Core(TM) i3-8100 CPU @ 3.60GHz | 1.2.0.475      | 64 Bit         | 3.43 GiB       | 3600 MHz       | ✔              |
+--------------------------------------------------+----------------+----------------------+-----------------------------------------+----------------+----------------+----------------+----------------

MySql is good choice for EF Core

In my case, I want to continue store data in CSV. Storage engine in MySql will help me do that and simply integrate my apps directly to data. It also helps to storage al data directly in memory for a cache. Of course, only InnoDB supports transactions, but I only consumer.

SHOW ENGINES

this command shows what  I have 9 store engines out of the box.

MEMORY    Hash based, stored in memory, useful for temporary tables    InnoDB    (DEFAULT)    Supports transactions, row-level locking, and foreign keys
PERFORMANCE_SCHEMA    Performance Schema
MyISAM    MyISAM storage engine
MRG_MYISAM    Collection of identical MyISAM tables
BLACKHOLE    YES    /dev/null storage engine (anything you write to it disappears)
CSV    CSV storage engine
ARCHIVE    Archive storage engine

It has workaround for enable storage engine in DataContext in EF code first solution using migration.

Solution:

• Using migrations (Instead of EnsureDatabaseCreated()) you can set innodb engine per db-session. (https://dev.mysql.com/doc/refman/5.7/en/storage-engine-setting.html)

protected override void Up(MigrationBuilder migrationBuilder)
{
migrationBuilder.Sql("SET default_storage_engine=INNODB");
//Another migration code is generated here
}

But then, you can generate  data with EF fluent interface code and only one command will be mode all this data to in memory database to fast performance calculations.

ALTER TABLE mytable ENGINE=MEMORY

Thank you, Oracle for this convenience.

It is look like a point in strategy where I can to upload normalized graph data from Arango DB to csv and then to the memory to load all data in GPU.

Audio is enough to study English

When I started to find good screen recording tool for Oracle Linux, I found only black screen recorded. But ffmpeg can record audio from game I can translate then to study.

There are steps to install ffmpeg

sudo dnf -y install https://download.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm

sudo dnf localinstall --nogpgcheck https://download1.rpmfusion.org/free/el/rpmfusion-free-release-8.noarch.rpm

sudo dnf install --nogpgcheck https://download1.rpmfusion.org/nonfree/el/rpmfusion-nonfree-release-8.noarch.rpm

sudo dnf install http://rpmfind.net/linux/epel/7/x86_64/Packages/s/SDL2-2.0.10-1.el7.x86_64.rpm 

sudo dnf install ffmpeg

sudo dnf -y install ffmpeg-devel

Then I have to run x11grab to record audio, command look like

ffmpeg -y \
-video_size 800x600 \
-framerate 30 -f x11grab -i :0.0 \
-f pulse -ac 2 -i default \
/tmp/screen1_recording_`date '+%Y-%m-%d_%H-%M-%S'`.mp4 \
&> /tmp/screen1_recording_`date '+%Y-%m-%d_%H-%M-%S'`.log 

Free music from Argentina

Every year i donated for archive.org because of a Common Creative music that lives there through decades. Its belongs to unknown authors, but within a years I listen selective albums day by day.

It is hard to find them on netlabels, catch with their covers, and hat I have for today:

[PSFR01] Quarter Bit - The Antarctica Files

Isla by Lost Tigres

Federico Palmolella - Keep on Groovin´ (ca455)

Jardín Crepe by NaveNube

Piacente & De Angelis - “Apud deum solaris I y II” (2019)

Alórea - Alórea (2017) (album) (CIOR-47)

They like another universe in music without brands and dictated motives by people.

Monodevelop on Oracle Linux

The time of old good tools is coming to my environment. I installed may be most usable code editor within a years by me in linux. This is a Monodevelop:

A script to install it on Oracle Linux 8 is:

sudo rpmkeys --import "http://pool.sks-keyservers.net/pks/lookup?op=get&search=0x3fa7e0328081bff6a14da29aa6a19b38d3d831ef"

su -c 'curl https://download.mono-project.com/repo/centos8-stable.repo | tee /etc/yum.repos.d/mono-centos8-stable.repo'

su -c 'curl https://download.mono-project.com/repo/centos8-vs.repo | tee /etc/yum.repos.d/mono-centos8-vs.repo'

sudo yum install mono-complete

sudo yum install gtk-sharp2

sudo yum install glib-sharp2

sudo yum install fsharp

sudo dnf install libglade2

sudo dnf install gtk-sharp2

install libssh2 from sources

sudo yum install monodevelop

Micro in chromeos shell

I get found my chrome brew stop install on my old Chromeos, but when I download install.sh from stable branch 3 years ago it installed again. Crew package manager I needed because of leak of standard command line editors. Crew installed micro, Ant and Gradle for me. I can use this smart editor for programming:

I wrote first part of book annotation tool in java directly in micro on my chromebook.

Application Inspector build with Roslyn analyze my AST parser

All are known, copy-paste is a bad thing for a code, but my project contains 95% of duplicated code. This code mostly generated with Roslyn to support ASt parser and importer for Arango graph db. When i run new Microsoft tool Application Inspector I finally know how large is auto generated code in my project.

This analyzer also know where I read from file, or even I need to switch from Newton.Json

I think it is security and discovery tool by Guy Acosta to better know what exactly you tried to run in your sandbox

Two languages and one on the host

It is easy to install Java in the ChromeOs, extracted from archive it simply works with sdk for development. Golang is supported by the crew as common package. .NET Core is hosted on the server for me, I can connect with ssh and Termius client.

Virtual Box 64bit setup

I tried to install Virtual Box from the rpm package but it did not work, because of a kernel module build. Here is an instruction from a link:

cd /etc/yum.repos.d/

wget http://download.virtualbox.org/virtualbox/rpm/rhel/virtualbox.repo

dnf install https://dl.fedoraproject.org/pub/epel/epel-release-latest-8.noarch.rpm

dnf install binutils gcc make patch libgomp glibc-headers glibc-devel kernel-headers kernel-devel dkms

dnf install VirtualBox-6.0

/usr/lib/virtualbox/vboxdrv.sh setup

usermod -a -G vboxusers user_name

Where Microsoft packages live

https://packages.microsoft.com is a home for packages like .net core and mssql-server. But it has other packages required for cloud maintenance, for example i found in the repo:

The Moby project
SCXCore
procdump

Not so much, but this is only one listed repo where i found .Net Core for my OS.