-
Notifications
You must be signed in to change notification settings - Fork 714
feat: add webUI to profiler #4544
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
base: main
Are you sure you want to change the base?
Conversation
Signed-off-by: hongkuanz <[email protected]>
hhzhang16
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can you include screenshots or a short demo in the MR description?
| AIPERF_PREFILL_ATTN_DP_NUM_REQ_RATIO = 4 | ||
|
|
||
| # Cost calculation defaults | ||
| GPU_COST_PER_HOUR = 3.0 # Cost per GPU per hour in dollars |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this consistent among all our GPUs?
| logger.addHandler(console_handler) | ||
|
|
||
| # Color palette for chart datasets | ||
| CHART_COLORS = [ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
if p_idx ever exceeds 8, the colors will be repeated; is that expected functionality?
| "#7f7f7f", # gray | ||
| ] | ||
|
|
||
| WEB_UI_SELECTION_TIMEOUT = 3600 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is a long timeout (1 hour), maybe it should be shorter
| data["prefill"]["chart"]["target_line"]["value"] = args.ttft | ||
| data["prefill"]["chart"]["target_line"]["label"] = f"Target TTFT: {args.ttft} ms" | ||
|
|
||
| data["decode"]["chart"]["target_line"]["value"] = args.itl | ||
| data["decode"]["chart"]["target_line"]["label"] = f"Target ITL: {args.itl} ms" |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is the UI user aware that we set defaults if they do not choose to set it?
parser.add_argument(
"--ttft",
type=float,
default=config.get("sla", {}).get("ttft", 50.0),
help="target Time To First Token (float, in milliseconds)",
)
parser.add_argument(
"--itl",
type=float,
default=config.get("sla", {}).get("itl", 10.0),
help="target Inter Token Latency (float, in milliseconds)",
)
| logger.info(f"Selection received: {plot_type}, row {row_idx}") | ||
|
|
||
| # Store selection for later confirmation | ||
| if plot_type == "cost": |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this stored locally/ephemerally?
Overview:
Details:
Where should the reviewer start?
Related Issues: (use one of the action keywords Closes / Fixes / Resolves / Relates to)