# Add the NVIDIA CUDA apt repository for Ubuntu 22.04
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/cuda-keyring_1.1-1_all.deb
sudo dpkg -i cuda-keyring_1.1-1_all.deb
sudo apt update

sudo tee /etc/apt/preferences.d/nvidia-cuda-pin <<'EOF'
Package: nvidia-dkms-open nvidia-firmware nvidia-kernel-common libnvidia-compute libnvidia-cfg1 libnvidia-decode libnvidia-gpucomp nvidia-persistenced
Pin: origin developer.download.nvidia.com
Pin-Priority: 1001
EOF

sudo apt install linux-headers-$(uname -r)
sudo apt install \
  nvidia-dkms-open \
  nvidia-firmware \
  nvidia-kernel-common \
  libnvidia-compute \
  libnvidia-cfg1 \
  libnvidia-decode \
  libnvidia-gpucomp \
  nvidia-persistenced

sudo dpkg -i --force-overwrite \
  /var/cache/apt/archives/nvidia-firmware_595.71.05-1ubuntu1_amd64.deb
sudo apt-get -f install

sudo dkms autoinstall
dkms status
# We expect something like this: nvidia/595.71.05, 6.8.0-117-generic, x86_64: installed

+-----------------------------------------------------------------------------------------+
| NVIDIA-SMI 595.71.05              Driver Version: 595.71.05      CUDA Version: 13.2     |
+-----------------------------------------+------------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id          Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |           Memory-Usage | GPU-Util  Compute M. |
|                                         |                        |               MIG M. |
|=========================================+========================+======================|
|   0  NVIDIA RTX PRO 6000 Blac...    On  |   00000000:16:00.0 Off |                  Off |
| 30%   31C    P8              7W /  300W |       2MiB /  97887MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
|   1  NVIDIA RTX PRO 6000 Blac...    On  |   00000000:34:00.0 Off |                  Off |
| 30%   31C    P8              4W /  300W |       2MiB /  97887MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
|   2  NVIDIA RTX PRO 6000 Blac...    On  |   00000000:AC:00.0 Off |                  Off |
| 30%   36C    P8             27W /  300W |       2MiB /  97887MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
|   3  NVIDIA RTX PRO 6000 Blac...    On  |   00000000:CA:00.0 Off |                  Off |
| 30%   33C    P8              7W /  300W |       2MiB /  97887MiB |      0%      Default |
|                                         |                        |                  N/A |
+-----------------------------------------+------------------------+----------------------+
+-----------------------------------------------------------------------------------------+
| Processes:                                                                              |
|  GPU   GI   CI              PID   Type   Process name                        GPU Memory |
|        ID   ID                                                               Usage      |
|=========================================================================================|
|  No running processes found                                                             |
+-----------------------------------------------------------------------------------------+

sudo nano /etc/default/grub  
# Change the line that start with `GRUB_CMDLINE_LINUX_DEFAULT`to:  
GRUB_CMDLINE_LINUX_DEFAULT="quiet splash pcie_aspm=off"
sudo update-grub

sudo tee /etc/modprobe.d/nvidia-power.conf <<EOF
options nvidia NVreg_DynamicPowerManagement=0x00
EOF

sudo apt install \
  nvidia-container-toolkit \
  nvidia-container-toolkit-base \
  libnvidia-container1 \
  libnvidia-container-tools

docker exec ollama ollama ps
NAME                                  ID              SIZE      PROCESSOR    CONTEXT    UNTIL
deepseek-r1:70b-llama-distill-fp16    7a7636f7c9fd    234 GB    100% GPU     131072     2 minutes from now

total duration:       2m2.821547202s
load duration:        22.803455278s
prompt eval count:    11 token(s)
prompt eval duration: 2.504074729s
prompt eval rate:     4.39 tokens/s
eval count:           1107 token(s)
eval duration:        1m36.827690145s
eval rate:            11.43 tokens/s

>>> Who is Donald E. Knuth?
Thinking...
Okay, so I need to figure out who Donald E. Knuth is. I've heard his name before in computer science circles, but I don't really know much about him. Let me start by
breaking down the question: "Who is Donald E. Knuth?"
First, I think he's a significant figure in computing because I remember seeing his name mentioned often in discussions about programming and algorithms. Maybe he wrote
some important books? Oh wait, I think he's known for "The Art of Computer Programming." That rings a bell. It sounds like a multi-volume set that covers a lot of topics
in computer science.
I also recall hearing about something called the Knuth-Morris-Pratt algorithm. So, if that's named after him, he must have contributed to algorithms. I believe that
algorithm is used for string searching, which is pretty fundamental in computer science. That makes me think he's done influential work in that area.
I'm not sure about his education or where he worked. Maybe he was affiliated with a well-known university? Stanford comes to mind because I know they have a strong CS
department, and I think Knuth might be associated there. Also, I've heard him referred to as a professor emeritus, which suggests he's retired but still respected in
academia.
There's also something about him offering rewards for finding errors in his books. I think it's called the "Knuth reward" or something like that. This shows his
commitment to accuracy and maybe a bit of humility, acknowledging that even experts can make mistakes.
What about his other contributions? I'm not certain, but he might have worked on typesetting systems. TeX and LaTeX are typesetting languages used widely in academia for
writing papers, especially in math and computer science. If Knuth created TeX, that's a huge contribution because it's been so influential in academic publishing.
He must have received numerous awards for his work. I can think of the Turing Award, which is like the Nobel Prize of computing. It wouldn't surprise me if he won that.
Also, perhaps other accolades like the National Medal of Science or honorary degrees from various universities.
His personality seems interesting too. He's known for his meticulousness and attention to detail, especially in his writing. I've heard that he uses a specific
typewriter or old computer systems because he prefers certain tools. This kind of dedication to his work probably contributes to the high quality of his publications.
I'm trying to remember if he has any notable quotes or philosophies. Something like "premature optimization is the root of all evil" comes to mind, though I'm not
entirely sure if that's him or someone else. If it is Knuth, then that reflects a design philosophy that emphasizes getting things right before optimizing them.
In terms of his influence on computer science education, his books must be foundational. They're probably used as textbooks in many courses and have shaped how
programming and algorithms are taught. His work has likely influenced generations of computer scientists and programmers.
I also think he's known for being quite approachable despite his prominence. Maybe he's active in corresponding with readers or engaging with the academic community.
This accessibility could contribute to his esteemed reputation.
So, putting it all together, Donald E. Knuth seems to be a polymath in computer science: an author of seminal books, a contributor to fundamental algorithms, creator of
essential tools like TeX, and a recipient of top awards. His work has had a lasting impact on both the theory and practice of computing.
...done thinking.
Donald E. Knuth is a renowned figure in computer science, celebrated for his profound contributions to the field. Here are the key aspects of his career and influence:
1. **Authorship**: He is best known for writing "The Art of Computer Programming," a multi-volume set that is a foundational text in computer science, covering a wide
range of topics with meticulous detail.
2. **Algorithms**: Knuth contributed significantly to algorithm design, notably the Knuth-Morris-Pratt algorithm, which is used for efficient string searching,
highlighting his impact on fundamental computing techniques.
3. **Academic Affiliation**: He is associated with Stanford University as a professor emeritus, reflecting his distinguished academic career and continued influence in
the field.
4. **Commitment to Accuracy**: Knuth offers monetary rewards for errors found in his books, demonstrating his dedication to precision and intellectual humility.
5. **Typesetting Contributions**: Creator of TeX and its extension Metafont, which revolutionized academic publishing, especially in mathematics and computer science,
through their precise typesetting capabilities.
6. **Awards and Honors**: A recipient of the Turing Award, among other prestigious accolades, recognizing his monumental contributions to computing.
7. **Philosophy and Personality**: Known for his meticulousness and design philosophy, encapsulated in quotes like "premature optimization is the root of all evil,"
emphasizing correctness over efficiency.
8. **Influence on Education**: His works have shaped computer science education, influencing generations through their use as textbooks and foundational resources.
9. **Accessibility**: Despite his prominence, Knuth is approachable and engages with the academic community, enhancing his reputation as a leader in the field.
In summary, Donald E. Knuth is a polymath whose contributions to algorithms, programming, typesetting, and education have left an indelible mark on computer science,
making him one of its most influential figures.