WEBVTT

00:00.000 --> 00:21.000
Hi, I'm Artichiro Seda from NTT. Today, I want to talk about the updates in DIMA page

00:21.000 --> 00:29.000
on 2.0. In this series, we are expanding the project focus to cover protecting AI, as well

00:29.000 --> 00:35.000
as Israeli running continuous on laptops.

00:35.000 --> 00:42.000
So, by the DIMA, RIMA stands for Linux Virtual Machine. It's optimized for running

00:42.000 --> 00:49.000
containers and AI agents, but early candidates for other workloads as well.

00:49.000 --> 00:54.000
And DIMA features automatic host-of-fire system sharing.

00:54.000 --> 01:03.000
That means the host-of-fire system can be easily shared to the case operating system.

01:03.000 --> 01:07.000
And it also supports automatic port routing.

01:07.000 --> 01:19.000
That means you can access to the TCP IP ports of the guest OS, just as a local host on the host operating system.

01:19.000 --> 01:28.000
So, you can just access the guest IP, just as a local host on the web browser running on the host operating system.

01:28.000 --> 01:38.000
And DIMA comes with a built-in integration for several functional times, including continuity, which is a default.

01:38.000 --> 01:44.000
And it also supports Joker, Potsumer, Kubernetes, and Aptiner.

01:44.000 --> 01:47.000
And DIMA supports several host operating systems.

01:47.000 --> 01:52.000
But in the case of Mac OS, you can just use PURU to install DIMA, PURU installed DIMA.

01:52.000 --> 01:57.000
And you can use DIMA CTR set to stretch our virtual machine.

01:57.000 --> 02:04.000
And you can just prefix as a DIMA command in the disk commands, such as not CTR.

02:04.000 --> 02:10.000
Not CTR means continuity CTR uses the whole controlling or continuity.

02:10.000 --> 02:15.000
Basically, it's similar to Joker.

02:15.000 --> 02:23.000
So, this is Mac OS, and you can just pre-painted DIMA to Union.

02:23.000 --> 02:26.000
And it's not Linux.

02:26.000 --> 02:35.000
And inside DIMA share, the host directly is a business goal.

02:35.000 --> 02:39.000
This is mounted on the Mac OS host operating system.

02:39.000 --> 02:50.000
And you can run continuous using another CTR command.

02:50.000 --> 02:57.000
This round is another CTR container, who running NX.

02:57.000 --> 03:04.000
And you can just access to the test port as a Joker host.

03:04.000 --> 03:09.000
So this is conducting through the disk as a Joker host.

03:09.000 --> 03:16.000
So basically, it just looks like a process running on the host operating system directly.

03:16.000 --> 03:28.000
But actually, it's wrapped inside the virtual machine.

03:28.000 --> 03:34.000
So, here is the project similar to DIMA.

03:34.000 --> 03:37.000
One example is WC2.

03:37.000 --> 03:43.000
So, with WC2, the Windows host 5 system is visible to the Linux disk.

03:43.000 --> 03:49.000
And the Linux disk supports are just accessible as a Joker host on the Windows host.

03:49.000 --> 03:57.000
So, it's a very similar to DIMA, but the WC2, of course, only supports Windows host.

03:58.000 --> 04:03.000
And it reminds also similar to big runs in some sense.

04:03.000 --> 04:05.000
But big runs is not property.

04:05.000 --> 04:11.000
And big runs also missing some features such as automatic port running.

04:11.000 --> 04:16.000
So you can just configure big runs to who are the ports.

04:16.000 --> 04:22.000
But it doesn't automatically scan ports used in the guest operating system.

04:22.000 --> 04:26.000
So it's quite different from DIMA in that sense.

04:26.000 --> 04:29.000
And also DIMA is a very similar to DIMA.

04:29.000 --> 04:33.000
But it's only made for Joker.

04:33.000 --> 04:37.000
And it's now abandoned a few years ago.

04:37.000 --> 04:42.000
And DIMA is also similar to the desktop, but without GUI.

04:42.000 --> 04:46.000
But the DIMA desktop is just made for the cover.

04:46.000 --> 04:51.000
Of course, and it's property.

04:51.000 --> 04:56.000
And let me talk about origin of the project in the current status.

04:56.000 --> 05:03.000
So I launched the project five years ago for the sake of promoting

05:03.000 --> 05:08.000
Continuity, including NRC to market or users.

05:08.000 --> 05:14.000
So this project was originally designed as Continuity in my scene.

05:14.000 --> 05:19.000
But through the course of the community, the scope has expanded.

05:19.000 --> 05:21.000
That's just for a quantity.

05:21.000 --> 05:24.000
So it's the Linux machine.

05:24.000 --> 05:26.000
Naturally, it's the machine.

05:26.000 --> 05:30.000
So we support several container engines, including Joker,

05:30.000 --> 05:32.000
Pokémon, Kubernetes, and Optiner.

05:32.000 --> 05:37.000
And we also support known container workloads.

05:37.000 --> 05:43.000
For example, it's also used for sandboxing AI coding engines.

05:43.000 --> 05:48.000
So you know, if you run AI coding agents like Clouds,

05:48.000 --> 05:50.000
it's a oxygen three rooms.

05:50.000 --> 05:54.000
Everything under your home directory, or maybe it's

05:54.000 --> 05:58.000
installs some random properties from the internet.

05:58.000 --> 06:04.000
So it's really scary to run AI coding agents, or you are up directly.

06:04.000 --> 06:08.000
So you can use the DIMA to send up such AI coding agents.

06:08.000 --> 06:11.000
And DIMA is also used for a way.

06:11.000 --> 06:18.000
You want to run a known Ubuntu OS such as Hydra on GitHub Actions.

06:18.000 --> 06:24.000
There was a GitHub Actions on this supports Ubuntu as a Linux guest.

06:24.000 --> 06:28.000
But you can just run DIMA inside GitHub Actions.

06:28.000 --> 06:35.000
So you can run any distribution such as Hydra, or

06:35.000 --> 06:40.000
Princess AI, or Work Linux, or Aromalyics, or Optine,

06:40.000 --> 06:45.000
and other distributions, and there's GitHub Actions.

06:45.000 --> 06:49.000
And DIMA is originally made of marketing tools,

06:49.000 --> 06:53.000
but we now support several host operating systems,

06:53.000 --> 06:56.000
including Linux Windows, new to BSD, and Dragon Privacy.

06:56.000 --> 06:59.000
We don't support previously because our queueing

06:59.000 --> 07:02.000
doesn't really work well on previously host.

07:02.000 --> 07:09.000
But I think we can see a real support previously as well.

07:09.000 --> 07:14.000
And we have lots of support, the free software projects

07:14.000 --> 07:16.000
based on DIMA.

07:16.000 --> 07:19.000
The most of the famous one is KORIMA.

07:19.000 --> 07:23.000
This is an alternative to the AI for DIMA,

07:23.000 --> 07:27.000
with a Docker at the default engine.

07:27.000 --> 07:30.000
KORIMA is very famous.

07:30.000 --> 07:34.000
Maybe it's more famous than DIMA itself.

07:34.000 --> 07:39.000
It has a very huge number of GitHub stores.

07:39.000 --> 07:47.000
So with KORIMA, you can just run Joker, or MacOS, very easily.

07:47.000 --> 07:53.000
And it doesn't need a proprietary Joker desktop license.

07:53.000 --> 07:58.000
And also there's a runger desktop license.

07:58.000 --> 08:04.000
It's a combination of DIMA and KORIMA, and KORIMA.

08:04.000 --> 08:08.000
So this is a screenshot of runger desktop.

08:08.000 --> 08:16.000
And this is a JOR dashboard, or you can check the new

08:16.000 --> 08:24.000
other status of DIMA runger desktop running KORIMA.

08:24.000 --> 08:27.000
And this is also free software.

08:27.000 --> 08:30.000
And the next run is a print.

08:30.000 --> 08:33.000
It's AWS product.

08:33.000 --> 08:38.000
And it's made for local development with AWS services,

08:38.000 --> 08:44.000
such as AWS server, application model,

08:44.000 --> 08:51.000
AWS, some of the recent way announced this support.

08:51.000 --> 08:58.000
And which is also free software as well.

08:58.000 --> 09:06.000
And there are projects like DIMA, GUI, which provides GUI based on

09:06.000 --> 09:10.000
huge framework, like this.

09:10.000 --> 09:14.000
And also, platform desktop supports

09:14.000 --> 09:16.000
managing DIMA instances.

09:16.000 --> 09:29.000
And we have more server-party projects.

09:29.000 --> 09:32.000
So let's talk about how it works.

09:32.000 --> 09:35.000
So this is a picture of DIMA.

09:35.000 --> 09:41.000
So DIMA is a DIMA city CIRI, KORIMA interface.

09:41.000 --> 09:47.000
This is used by Shuman, or GUI, or MCP programming.

09:47.000 --> 09:55.000
And this CIRI can be used for managing DIMA host-egent processes.

09:55.000 --> 10:00.000
And host-egent processes use this BIM driver

10:00.000 --> 10:05.000
to support a server-party machine implementations.

10:05.000 --> 10:10.000
And this driver runs this BIM, and also connects to

10:10.000 --> 10:12.000
the network driver.

10:12.000 --> 10:16.000
And instead of BIM, we have DIMA-egent processes,

10:16.000 --> 10:22.000
which handles five-system sharing and port-wadding,

10:22.000 --> 10:33.000
and thinking real-time cook, and other jobs.

10:33.000 --> 10:36.000
And DIMA is designed to be

10:36.000 --> 10:41.000
incredible to support different implementations of

10:41.000 --> 10:47.000
hyperbysers, such as QEMU, or Apple's

10:47.000 --> 10:49.000
virtualization framework.

10:49.000 --> 10:51.000
And then was busy on MacOS.

10:51.000 --> 10:55.000
We also support WSR2 on Windows hosts.

10:55.000 --> 11:01.000
And in button 2.0, we added support for KLONKIT,

11:01.000 --> 11:05.000
which supports GPU acceleration on MacOS.

11:05.000 --> 11:10.000
And in DIMA-egent 2, we also support GRPC

11:10.000 --> 11:16.000
programming to support your own favorite hyperbysers

11:16.000 --> 11:18.000
implementation.

11:18.000 --> 11:24.000
And we also have the concept of inter-on-arm binary

11:24.000 --> 11:25.000
effectors.

11:25.000 --> 11:29.000
So the default is two-emium user mode.

11:29.000 --> 11:33.000
So you can just run inter-bying ways on

11:33.000 --> 11:37.000
the term instances of QEMU.

11:37.000 --> 11:41.000
And on MacOS hosts, you can also use ROS-8.2,

11:41.000 --> 11:45.000
which is more fast, faster than QEMU.

11:45.000 --> 11:51.000
And for five-system sharing, we support

11:51.000 --> 11:54.000
Part-VIOFS and Part-VIO90.

11:54.000 --> 11:58.000
And we also support reverse SSHFS,

11:58.000 --> 12:03.000
where the guest operating system doesn't support

12:03.000 --> 12:07.000
Part-VIOFS and Part-VIO90.

12:07.000 --> 12:12.000
And for networking, the default is user mode

12:12.000 --> 12:15.000
network based on GByser.

12:15.000 --> 12:18.000
So this company works in the user space,

12:18.000 --> 12:23.000
so it doesn't need any route privilege on the host.

12:23.000 --> 12:28.000
But when you want to access the virtual machine by IP,

12:28.000 --> 12:31.000
not just by the Roka host address,

12:31.000 --> 12:35.000
you can use a sorted BEMU driver.

12:35.000 --> 12:37.000
This needs route privilege.

12:37.000 --> 12:41.000
So you have to launch it with SIDO.

12:41.000 --> 12:45.000
And for busy driver, you can automatically use

12:45.000 --> 12:49.000
a busy network driver to access the BEMU by IP.

12:49.000 --> 12:55.000
This one only works with busy but this doesn't need SIDO command.

12:55.000 --> 12:58.000
And for port routing,

12:58.000 --> 13:03.000
we watch EVP events.

13:03.000 --> 13:08.000
It's a need to link SOC diagnosis events.

13:08.000 --> 13:15.000
And we use EVP to watch this port routing events.

13:15.000 --> 13:18.000
But this implementation doesn't work

13:18.000 --> 13:20.000
for key brand service ports,

13:20.000 --> 13:25.000
because key brand service ports are quite different from regular ports,

13:25.000 --> 13:29.000
because it uses IP tables.

13:29.000 --> 13:33.000
So we also have a custom logic for

13:33.000 --> 13:38.000
what's the key brand service ports.

13:38.000 --> 13:41.000
And the default,

13:41.000 --> 13:44.000
this row of the guestways is Ubuntu,

13:44.000 --> 13:47.000
but you can also choose other distribution

13:47.000 --> 13:50.000
as automatically used by our Linux,

13:50.000 --> 13:52.000
sent of stream, db and open services,

13:52.000 --> 13:55.000
Oracle Linux, and from Linux.

13:55.000 --> 13:57.000
And the default contenting is contenting,

13:57.000 --> 14:01.000
but you can also use updating a dog or a partner.

14:01.000 --> 14:04.000
In route risk mode by default,

14:04.000 --> 14:06.000
or in route for mode,

14:06.000 --> 14:10.000
when you need privileged operations.

14:10.000 --> 14:13.000
And you can also use several

14:13.000 --> 14:15.000
additional instructions,

14:15.000 --> 14:19.000
such as Kubernetes.

14:19.000 --> 14:23.000
And you can just choose your templates

14:23.000 --> 14:28.000
using this DMACC to your start command.

14:28.000 --> 14:34.000
And let me talk about recent updates.

14:34.000 --> 14:39.000
So we recently got promoted to CNC incorporating projects.

14:39.000 --> 14:43.000
CNC is responsible for creating a project,

14:43.000 --> 14:46.000
which is hosting several projects,

14:46.000 --> 14:48.000
such as Kubernetes.

14:48.000 --> 14:51.000
And CNC has three strategies.

14:51.000 --> 14:54.000
Sandbox, integrating and graduated.

14:54.000 --> 14:59.000
We joined CNC as a Sandbox in 2022,

14:59.000 --> 15:04.000
and we got promoted to integrating

15:04.000 --> 15:06.000
last year.

15:06.000 --> 15:10.000
And we anticipate that we can promote it

15:10.000 --> 15:16.000
to graduate to be able in the next couple of years.

15:16.000 --> 15:22.000
And the number of retailers is growing.

15:22.000 --> 15:27.000
So we now have more than 20,000 upstairs,

15:27.000 --> 15:33.000
and we have more than 170 contributors.

15:33.000 --> 15:42.000
So thanks to all the contributors and all the users.

15:42.000 --> 15:47.000
And in November, we released a remote version to 0.

15:47.000 --> 15:50.000
And this week, we added a product infrastructure

15:50.000 --> 15:54.000
to auto implementing new features

15:54.000 --> 15:56.000
without modifying remote.

15:56.000 --> 15:58.000
So we have the concept of BM driver

15:58.000 --> 16:03.000
projects to support additional hyperbysers.

16:03.000 --> 16:09.000
And we also support CLI projects to add more subcommands

16:09.000 --> 16:11.000
to the DMACC to CLI.

16:11.000 --> 16:15.000
And we also have projects for URL schema

16:15.000 --> 16:20.000
for fixing templates from custom remote addresses.

16:20.000 --> 16:25.000
And we support GP access additional using

16:25.000 --> 16:27.000
care on to the BM driver.

16:27.000 --> 16:30.000
And we also added MCP server,

16:30.000 --> 16:35.000
model contest protocol server for protecting AI agents.

16:35.000 --> 16:40.000
So in just weeks, we are extending things

16:40.000 --> 16:44.000
to cover AI as well as consumers.

16:44.000 --> 16:49.000
So the original goal in 2021 was to facilitate

16:49.000 --> 16:52.000
running continuity on Mac OS.

16:52.000 --> 16:57.000
But the project turned out to be very useful

16:57.000 --> 17:00.000
for securing AI agents too.

17:00.000 --> 17:05.000
So as to prevent them from accessing the host of files

17:05.000 --> 17:06.000
and the commands.

17:06.000 --> 17:10.000
So you know, AI agents may make

17:10.000 --> 17:15.000
hallucination to remove files under the home directory.

17:15.000 --> 17:18.000
So this example is from Reddit,

17:18.000 --> 17:23.000
but this example quote attempted

17:23.000 --> 17:27.000
to clean up some packages and unused files.

17:27.000 --> 17:31.000
And quote launched RMR commands

17:31.000 --> 17:34.000
to test, test, test, test, press, press.

17:34.000 --> 17:38.000
And the program is that it even removes

17:38.000 --> 17:41.000
the home directory to the source.

17:41.000 --> 17:46.000
So it removes everything in the home directory.

17:46.000 --> 17:51.000
So it's very scary to run AI on your laptop

17:51.000 --> 17:52.000
directly.

17:52.000 --> 17:57.000
So it's useful for securing such AI agents.

17:57.000 --> 18:02.000
And I also have to note that AI may also make

18:02.000 --> 18:06.000
hallucination to install fake packages with

18:06.000 --> 18:10.000
plausible names by running people in store or

18:10.000 --> 18:15.000
MPM install or go install or party buyers.

18:15.000 --> 18:20.000
And AI agents now have a web search too.

18:20.000 --> 18:26.000
And AI agents are sometimes deceived by fake websites

18:26.000 --> 18:30.000
once they're internet, appears the web search too

18:30.000 --> 18:37.000
to install some malicious purchases.

18:37.000 --> 18:42.000
So AI agents often come with built in a sandboxing

18:42.000 --> 18:44.000
who is an example.

18:44.000 --> 18:49.000
They use a dandroque or joker container or Linux,

18:49.000 --> 18:53.000
but it's not as strong as virtual machines.

18:53.000 --> 18:59.000
And some AI agents use a sandbox exec,

18:59.000 --> 19:03.000
which is similar to dandroque or macOS,

19:03.000 --> 19:06.000
but this sandbox exec command has been

19:06.000 --> 19:11.000
deported almost a decade ago.

19:11.000 --> 19:14.000
And Apple recommends using up sandbox,

19:14.000 --> 19:18.000
but it's not direct refreshment to a sandbox exec

19:18.000 --> 19:19.000
command.

19:19.000 --> 19:22.000
So AI agent products are still

19:22.000 --> 19:26.000
derived on sandbox exec commands.

19:26.000 --> 19:31.000
So we can use Dima as a reversal sandbox

19:31.000 --> 19:34.000
for any AI agent.

19:34.000 --> 19:38.000
And it's also, it can coexist with

19:38.000 --> 19:45.000
the built in sandboxing provided by AI agents.

19:45.000 --> 19:49.000
And we're running AI with Dima,

19:49.000 --> 19:51.000
there are two scenarios.

19:51.000 --> 19:54.000
AI inside Dima and AI outside Dima.

19:54.000 --> 19:57.000
So AI inside Dima is running codex,

19:57.000 --> 19:59.000
compiled or crawled or jiminite,

19:59.000 --> 20:02.000
or maybe open code inside Dima.

20:02.000 --> 20:07.000
And you can also do an inference inside Dima

20:07.000 --> 20:11.000
using GPU acceleration.

20:11.000 --> 20:17.000
And AI outside Dima means AI agent running on the host

20:17.000 --> 20:21.000
can connect to Dima and Dima provides

20:21.000 --> 20:26.000
MCP servers for protecting five system course

20:26.000 --> 20:30.000
and command execution course.

20:30.000 --> 20:34.000
And you can also use the escorts and remote SSH

20:34.000 --> 20:39.000
and compile notes with Dima.

20:39.000 --> 20:49.000
And we have example of running AI inside Dima.

20:49.000 --> 20:52.000
For example, you can use this Dima CTR

20:52.000 --> 20:55.000
start command to only mouse working

20:55.000 --> 20:57.000
directly in mid-right mode.

20:57.000 --> 21:00.000
And we can install open code.

21:00.000 --> 21:04.000
So I don't have time to show the demo.

21:04.000 --> 21:09.000
But inside websites, we have example for continuous.

21:09.000 --> 21:13.000
And we also have examples for AI agents.

21:13.000 --> 21:17.000
So you can choose a crawled code or codex or jiminite or

21:17.000 --> 21:20.000
jiminite and compile.

21:20.000 --> 21:26.000
And you can just run its inside Dima.

21:26.000 --> 21:30.000
And we also have GPU acceleration.

21:30.000 --> 21:34.000
So that must CPP running inside Dima

21:34.000 --> 21:39.000
uses a partner to talk to a partner or GPU.

21:39.000 --> 21:41.000
And the partner or GPU,

21:41.000 --> 21:46.000
you talk to both Dima and Brieke,

21:46.000 --> 21:50.000
which translates partner library code

21:50.000 --> 21:56.000
to meta library course, which is used by Apple.

21:56.000 --> 22:07.000
I think I can show a quick demo.

22:07.000 --> 22:13.000
So this is using a partner GPU,

22:13.000 --> 22:17.000
a partner or GPU binas, a point for max.

22:17.000 --> 22:25.000
And I can talk to AI running inside Dima.

22:25.000 --> 22:32.000
Yeah, so the performance is not so much.

22:33.000 --> 22:36.000
And for running AI outside Dima,

22:36.000 --> 22:39.000
Dima exposes several MCP tools,

22:39.000 --> 22:43.000
such as district where you apply run share command.

22:43.000 --> 22:46.000
These are a similar to jiminite CRIs commands,

22:46.000 --> 22:49.000
but it's stored with some of us using BM.

22:49.000 --> 22:53.000
And we plan to raise budget to a point to run in March.

22:53.000 --> 22:56.000
And in this race, we have single modes,

22:56.000 --> 22:57.000
unlike months.

22:57.000 --> 22:59.000
I think the diverse are with Dima,

22:59.000 --> 23:01.000
but only after user information.

23:01.000 --> 23:04.000
So this prevents AI from the situation

23:04.000 --> 23:06.000
where a story I removed everything,

23:06.000 --> 23:08.000
including Dr. Dima.

23:08.000 --> 23:12.000
So this should be highly useful running AI inside Dima.

23:12.000 --> 23:15.000
And I don't have time to cover our future ideas,

23:15.000 --> 23:20.000
but we've come contributions in this field.

23:20.000 --> 23:23.000
And we have a huge community,

23:23.000 --> 23:25.000
and we have website, GitHub, and Slack, and Twitter,

23:26.000 --> 23:27.000
and thank you.

23:38.000 --> 23:39.000
Any questions?

23:46.000 --> 23:47.000
Sorry?

23:47.000 --> 23:48.000
Any questions?

23:48.000 --> 23:51.000
So Dima, on different posters, talking to each other,

23:51.000 --> 23:53.000
or they have to run on the same post?

23:54.000 --> 23:59.000
So you can use a bridge device

23:59.000 --> 24:02.000
to directly connect virtual machine

24:02.000 --> 24:04.000
to the host network.

24:04.000 --> 24:08.000
So you can just connect multiple instances

24:08.000 --> 24:11.000
running on several posts.

24:23.000 --> 24:28.000
How does Dima remove this mouse and try to?

24:28.000 --> 24:30.000
What have we received too?

24:30.000 --> 24:32.000
In the object itself, you're lost.

24:32.000 --> 24:34.000
The C-try is mounted,

24:34.000 --> 24:36.000
between two of the VMDU terms.

24:36.000 --> 24:39.000
How does Dima remove the mouse and feed?

24:39.000 --> 24:42.000
Actually, I don't use WC2 by myself,

24:42.000 --> 24:46.000
so I'm not sure about WC2 implementation.

24:47.000 --> 24:49.000
Please, don't worry.

24:51.000 --> 24:53.000
These are a lot of big questions.

24:53.000 --> 24:54.000
That's it.