forked from opea-project/Enterprise-Inference
-
Notifications
You must be signed in to change notification settings - Fork 1
Expand file tree
/
Copy pathibm_catalog.json
More file actions
executable file
·555 lines (552 loc) · 25.2 KB
/
ibm_catalog.json
File metadata and controls
executable file
·555 lines (552 loc) · 25.2 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
{
"products": [
{
"label": "Intel® AI for Enterprise Inference",
"name": "da-enterprise-inference",
"version": "1.3.1",
"product_kind": "solution",
"tags": [
"opea",
"intel_gaudi",
"enterprise-inference"
],
"keywords": [
"opea",
"intel_gaudi",
"enterprise-inference"
],
"short_description": "Intel® AI for Enterprise Inference (Enterprise Inference) is aimed to streamline the deployment and management of inference services on Intel® hardware.",
"long_description": "This solution automates the provisioning of virtual machines on IBM Cloud using Terraform, optimized for AI workloads powered by Intel® Gaudi® 3 AI accelerators. It deploys a Kubernetes cluster that supports both single-node and multi-node configurations. The multi-node setup includes three control plane nodes for orchestration and cluster management, along with worker nodes equipped with Intel® Gaudi® 3 for high-performance AI inference.The deployment integrates the Intel® AI for Enterprise Inference framework from the Open Platform for Enterprise AI (OPEA), enabling scalable deployment of Large Language Models (LLMs) across various configurations. It also includes key enterprise-grade features such as GenAI Gateway for secure and scalable access to generative AI models, Token Telemetry for detailed tracking and analysis of token usage across inference workloads, and Observability for real-time monitoring of system performance, resource utilization, and model behavior.By combining cutting-edge AI infrastructure with cloud automation, this solution provides a streamlined and efficient approach to deploying and managing machine learning models at scale.\n\n**Prerequisites:**\n- Review the [Getting Started Guide](https://github.com/opea-project/Enterprise-Inference/blob/main/third_party/IBM/docs/getting-started.md) for deployment instructions\n- Consult the [Sizing Guide](https://github.com/opea-project/Enterprise-Inference/blob/main/third_party/IBM/docs/sizing-guide.md) to understand Intel® Gaudi® 3 AI Accelerator capabilities and choose the optimal configuration for your AI models",
"features": [
{
"title": "Optimized AI Deployment",
"description": "Leverage Intel® Gaudi® 3 AI accelerators to maximize performance for AI workloads, ensuring efficient processing and model deployment"
},
{
"title": "Automated Infrastructure Setup",
"description": "Utilization of Terraform scripts to automate the provisioning of virtual server instances (VSI) on IBM Cloud, reducing setup time and minimizing manual intervention"
},
{
"title": "Seamless Integration with OPEA's Intel® AI for Enterprise Inference",
"description": "Deploy large language models (LLMs) using OPEA's Intel AI for Enterprise Inference framework, streamlining operations and improving productivity"
}
],
"flavors": [
{
"label": "QuickStart",
"name": "quickstart",
"working_directory": "third_party/IBM/patterns/quickstart",
"index": 1,
"compliance": {},
"architecture": {},
"install_type": "fullstack",
"configuration": [
{
"key": "ssh_key",
"type": "string",
"default_value": "ssh_key",
"description": "This variable specifies the ssh key.",
"required": true
},
{
"key": "ssh_private_key",
"type": "multiline_secure_value",
"display_name": "ssh_private_key",
"default_value": "-----BEGIN OPENSSH PRIVATE KEY-----",
"description": "This variable specifies the ssh key.",
"required": true,
"custom_config": {
"grouping": "deployment",
"original_grouping": "deployment",
"type": "multiline_secure_value"
}
},
{
"key": "ibmcloud_region",
"type": "string",
"default_value": "",
"description": "IBM Cloud Region",
"required": true,
"options": [
{
"displayname": "Dallas (us-south)",
"value": "us-south"
},
{
"displayname": "Washington DC (us-east)",
"value": "us-east"
},
{
"displayname": "Frankfurt (eu-de)",
"value": "eu-de"
}
]
},
{
"key": "ibmcloud_api_key",
"type": "multiline_secure_value",
"display_name": "Multiline secure value",
"required": true,
"custom_config": {
"grouping": "deployment",
"original_grouping": "deployment",
"type": "multiline_secure_value"
}
},
{
"key": "instance_zone",
"type": "string",
"default_value": "",
"description": "This variable specifies the availability zone where the instance will be created",
"required": true,
"options": [
{
"displayname": "eu-de-1",
"value": "eu-de-1"
},
{
"displayname": "us-south-2",
"value": "us-south-2"
},
{
"displayname": "us-east-3",
"value": "us-east-3"
},
{
"displayname": "us-east-2",
"value": "us-east-2"
}
]
},
{
"key": "cluster_url",
"type": "string",
"default_value": "api.example.com",
"description": "Enter the domain name or url of the cluster",
"required": true
},
{
"key": "models",
"type": "string",
"default_value": "",
"description": "Choose the pre-tested model from the dropdown to deploy.",
"required": true,
"options": [
{
"displayname": "meta-llama/Llama-3.1-8B-Instruct (1 card,disk-20GB)",
"value": "1"
},
{
"displayname": "meta-llama/Llama-3.1-405B-Instruct (8 card ,disk-900GB)",
"value": "3"
},
{
"displayname": "meta-llama/Llama-3.3-70B-Instruct (4 card ,disk-150GB)",
"value": "4"
},
{
"displayname": "meta-llama/Llama-4-Scout-17B-16E-Instruct (4 card ,disk-200GB)",
"value": "5"
},
{
"displayname": "Qwen/Qwen2.5-32B-Instruct (1 card ,disk-70GB)",
"value": "6"
}
]
},
{
"key": "hugging_face_token",
"type": "multiline_secure_value",
"default_value": "__NULL__",
"description": "This variable specifies the Hugging Face token. You can also create a token by referring to the documentation at https://huggingface.co/docs/hub/en/security-tokens. ",
"required": true
},
{
"key": "instance_name",
"type": "string",
"default_value": "__NULL__",
"description": "This variable specifies the name of the instance to be created",
"required": true
},
{
"key": "vpc",
"type": "string",
"default_value": "__NULL__",
"description": "This variable specifies vpc where all the resources will be created",
"required": true
},
{
"key": "security_group",
"type": "string",
"default_value": "__NULL__",
"description": "This variable specifies security group used by the resources",
"required": true
},
{
"key": "public_gateway",
"type": "string",
"default_value": "__NULL__",
"description": "This variable specifies public gateway used by the resources",
"required": true
},
{
"key": "subnet",
"type": "string",
"default_value": "__NULL__",
"description": "This variable specifies subnet used by the resources",
"required": true
},
{
"key": "resource_group",
"default_value": "Default",
"description": "This variable lists resource group name used by the resources",
"required": true,
"custom_config": {
"config_constraints": {
"identifier": "rg_name"
},
"grouping": "deployment",
"original_grouping": "deployment",
"type": "resource_group"
}
},
{
"key": "user_cert",
"type": "multiline_secure_value",
"display_name": "Full chain Certificate",
"default_value": "-----BEGIN CERTIFICATE-----",
"description": "The contents of the TLS certificate (PEM format)",
"required": true
},
{
"key": "user_key",
"type": "multiline_secure_value",
"display_name": "Certificate key",
"default_value": "-----BEGIN PRIVATE KEY-----",
"description": "The contents of the TLS private key (PEM format)",
"required": true
},
{
"key": "deployment_mode",
"type": "string",
"display_name": "Deployment Mode",
"default_value": "single-node",
"description": "Deployment mode for the infrastructure (single-node or multi-node)",
"required": true,
"options": [
{
"displayname": "Single Node",
"value": "single-node"
},
{
"displayname": "Multi Node",
"value": "multi-node"
}
]
},
{
"key": "control_plane_count",
"type": "number",
"display_name": "Control plane nodes count",
"default_value": "3",
"description": "Number of control plane nodes (1 for single or 3 for HA) - only used in multi-node mode",
"required": true
},
{
"key": "worker_gaudi_count",
"type": "number",
"display_name": "Worker nodes count",
"default_value": "1",
"description": "Number of Gaudi worker nodes for inference - only used in multi-node mode",
"required": true
},
{
"key": "control_plane_names",
"type": "list(string)",
"display_name": "Control plane names",
"default_value": "[]",
"description": "Custom names for control plane nodes. If not provided, defaults to 'inference-control-plane-01', etc."
},
{
"key": "worker_gaudi_names",
"type": "list(string)",
"display_name": "Gaudi worker names",
"default_value": "[]",
"description": "Custom names for Gaudi worker nodes. If not provided, defaults to 'inference-workload-gaudi-node-01', etc."
},
{
"key": "instance_profile",
"type": "string",
"default_value": "gx3d-160x1792x8gaudi3",
"description": "This variable specifies the profile using which the instance will be created"
},
{
"key": "cert_path",
"type": "string",
"default_value": "~/certs/cert.pem",
"description": "The path to the certificate file"
},
{
"key": "key_path",
"type": "string",
"default_value": "~/certs/key.pem",
"description": "The path to the key file"
},
{
"key": "cpu_or_gpu",
"type": "string",
"default_value": "gaudi3",
"description": "This variable specifies where the model should be running"
},
{
"key": "vault_pass_code",
"type": "multiline_secure_value",
"default_value": "",
"description": "Enter the master vault pass code",
"required": true
},
{
"key": "deploy_kubernetes_fresh",
"type": "string",
"default_value": "yes",
"description": "This variable specfies whether to deploy Kubernetes cluster freshly"
},
{
"key": "deploy_ingress_controller",
"type": "string",
"default_value": "yes",
"description": "This variable specfies whether to deploy NGNIX ingress controller or not"
},
{
"key": "deploy_genai_gateway",
"type": "string",
"default_value": "yes",
"description": "This variable specfies whether we need to deploy Gen AI Gateway"
},
{
"key": "deploy_observability",
"type": "string",
"default_value": "no",
"description": "This variable specfies whether we need to run observability"
},
{
"key": "deploy_llm_models",
"type": "string",
"default_value": "no",
"description": "This variable specfies whether we need to deploy LLM models"
}
]
},
{
"label": "Standard",
"name": "standard",
"working_directory": "third_party/IBM/patterns/standard",
"compliance": {},
"index": 2,
"architecture": {},
"install_type": "fullstack",
"configuration": [
{
"key": "ssh_key",
"type": "string",
"default_value": "ssh_key",
"description": "This variable specifies the ssh key.",
"required": true
},
{
"key": "ssh_private_key",
"type": "multiline_secure_value",
"display_name": "ssh_private_key",
"default_value": "-----BEGIN OPENSSH PRIVATE KEY-----",
"description": "This variable specifies the ssh key.",
"required": true,
"custom_config": {
"grouping": "deployment",
"original_grouping": "deployment",
"type": "multiline_secure_value"
}
},
{
"key": "ibmcloud_region",
"type": "string",
"default_value": "",
"description": "IBM Cloud Region",
"required": true,
"options": [
{
"displayname": "Dallas (us-south)",
"value": "us-south"
},
{
"displayname": "Washington DC (us-east)",
"value": "us-east"
},
{
"displayname": "Frankfurt (eu-de)",
"value": "eu-de"
}
]
},
{
"key": "ibmcloud_api_key",
"type": "multiline_secure_value",
"display_name": "Multiline secure value",
"required": true,
"custom_config": {
"grouping": "deployment",
"original_grouping": "deployment",
"type": "multiline_secure_value"
}
},
{
"key": "instance_zone",
"type": "string",
"default_value": "",
"description": "This variable specifies the availability zone where the instance will be created",
"required": true,
"options": [
{
"displayname": "eu-de-1",
"value": "eu-de-1"
},
{
"displayname": "us-south-2",
"value": "us-south-2"
},
{
"displayname": "us-east-3",
"value": "us-east-3"
},
{
"displayname": "us-east-2",
"value": "us-east-2"
}
]
},
{
"key": "cluster_url",
"type": "string",
"default_value": "api.example.com",
"description": "Enter the domain name or url of the cluster",
"required": true
},
{
"key": "models",
"type": "string",
"default_value": "",
"description": "Choose the pre-tested model from the dropdown to deploy.",
"required": true,
"options": [
{
"displayname": "meta-llama/Llama-3.1-8B-Instruct (1 card,disk-20GB)",
"value": "1"
},
{
"displayname": "meta-llama/Llama-3.1-405B-Instruct (8 card ,disk-900GB)",
"value": "3"
},
{
"displayname": "meta-llama/Llama-3.3-70B-Instruct (4 card ,disk-150GB)",
"value": "4"
},
{
"displayname": "meta-llama/Llama-4-Scout-17B-16E-Instruct (4 card ,disk-200GB)",
"value": "5"
},
{
"displayname": "Qwen/Qwen2.5-32B-Instruct (1 card ,disk-70GB)",
"value": "6"
}
]
},
{
"key": "hugging_face_token",
"type": "multiline_secure_value",
"default_value": "__NULL__",
"description": "This variable specifies the Hugging Face token. You can also create a token by referring to the documentation at https://huggingface.co/docs/hub/en/security-tokens. ",
"required": true
},
{
"key": "resource_group",
"default_value": "Default",
"description": "This variable lists resource group name used by the resources",
"required": true,
"custom_config": {
"config_constraints": {
"identifier": "rg_name"
},
"grouping": "deployment",
"original_grouping": "deployment",
"type": "resource_group"
}
},
{
"key": "user_cert",
"type": "multiline_secure_value",
"display_name": "Full chain Certificate",
"default_value": "-----BEGIN CERTIFICATE-----",
"description": "The contents of the TLS certificate (PEM format)",
"required": true
},
{
"key": "user_key",
"type": "multiline_secure_value",
"display_name": "Certificate key",
"default_value": "-----BEGIN PRIVATE KEY-----",
"description": "The contents of the TLS private key (PEM format)",
"required": true
},
{
"key": "ssh_allowed_cidr",
"type": "string",
"default_value": "\"\"",
"description": "CIDR block(s) allowed for SSH, HTTP, and HTTPS access. Enter a single CIDR or comma-separated list like: 198.0.0.0/8,192.0.0.0/8,169.0.0.0/8. Change to your organization's IP range for better security. For development/testing use 0.0.0.0/0",
"required": true
},
{
"key": "deployment_mode",
"type": "string",
"display_name": "Deployment Mode",
"default_value": "single-node",
"description": "Deployment mode for the infrastructure (single-node or multi-node)",
"required": true,
"options": [
{
"displayname": "Single Node",
"value": "single-node"
},
{
"displayname": "Multi Node",
"value": "multi-node"
}
]
},
{
"key": "control_plane_count",
"type": "number",
"display_name": "Control plane nodes count",
"default_value": "3",
"description": "Number of control plane nodes (1 for single or 3 for HA) - only used in multi-node mode",
"required": true
},
{
"key": "worker_gaudi_count",
"type": "number",
"display_name": "Worker nodes count",
"default_value": "1",
"description": "Number of Gaudi worker nodes for inference - only used in multi-node mode",
"required": true
},
{
"key": "vault_pass_code",
"type": "multiline_secure_value",
"default_value": "",
"description": "Enter the master vault pass code",
"required": true
}
]
}
]
}
]
}