Meta tags:
description= Create HPC instances with Cloud RDMA in bulk to use reserved blocks of resources in dense deployments.;
Headings (most frequently used words):
create, gcloud, rest, bulk, instances, with, and, optional, required, reservation, bound, spot, hpc, optimized, h4d, stay, organized, collections, save, categorize, content, based, on, your, preferences, before, you, begin, overview, virtual, private, cloud, networks, placement, policy, vm, in, what, next, console, roles, instruction, guides, script, products, pricing, support, resources, engage, permissions,
Text of the page (most frequently used words):
the (447), #create (187), for (183), and (152), vms (136), network (119), with (105), use (103), you (102), compute (82), #instances (74), about (73), instance (68), troubleshoot (62), cloud (56), disk (56), using (52), that (51), manage (48), reservation (48), interface (47), type (47), view (44), machine (44), following (42), placement (40), name (40), disks (40), image (39), server (39), set (38), configure (38), google (36), h4d (36), irdma (36), address (35), gvnic (35), region (34), mig (34), overview (33), windows (33), policy (32), bulk (32), subnet (31), performance (31), vpc (30), gvnic_name_prefix (30), example (29), engine (29), from (28), images (27), compact (26), external (26), can (26), want (26), specify (26), see (25), linux (25), optional (25), your (25), hyperdisk (25), one (24), available (24), gcloud (24), policies (23), project (23), networks (23), access (21), sql (21), regions (20), number (20), add (20), monitor (20), request (19), when (19), resources (18), creating (18), default (18), replace (18), creation (18), boot (18), zone (18), zones (18), configuration (18), ssh (18), custom (18), gpu (18), this (17), are (17), cluster (17), block (17), host (17), virtual (17), reservations (17), rdma_name_prefix (16), must (16), choose (16), count (16), snapshots (16), groups (16), rdma (15), connect (15), net (15), sub (15), delete (15), apply (15), management (15), recommendations (15), information (14), optimized (14), persistent (14), best (14), hpc (14), workload (14), load (14), ipv4 (13), then (13), interfaces (13), shutdown (13), mode (13), spot (13), start (13), command (13), resource (13), microsoft (13), ipv4_only (12), projects (12), application (12), types (12), nic (12), storage (12), sole (12), availability (12), regional (12), have (11), policy_name (11), capacity (11), run (11), maintenance (11), metadata (11), based (11), practices (11), workloads (11), stateful (11), prefix (10), instead (10), parameter (10), list (10), supported (10), highmem (10), pattern (10), addresses (10), internal (10), existing (10), install (10), migs (10), machines (10), attached (10), migrate (10), logs (10), events (9), rest (9), used (9), family (9), operating (9), size (9), consumption (9), ipv6 (9), flag (9), falcon (9), rules (9), roles (9), service (9), snapshot (9), update (9), high (9), licenses (9), managed (9), tenant (9), multiple (9), openshift (9), time (9), action (8), samples (8), other (8), license (8), scale (8), external_ipv4_address (8), subnetwork (8), instanceproperties (8), resourcepolicies (8), project_id (8), script (8), image_family (8), image_project (8), disk_size (8), machine_type (8), name_pattern (8), which (8), where (8), specific (8), flags (8), value (8), place (8), group (8), console (8), uses (8), networking (8), permissions (8), authentication (8), enable (8), import (8), cpu (8), login (8), optimize (8), disaster (8), recovery (8), agent (8), guest (8), keys (8), flexibility (8), dns (8), tpu (8), code (7), thumb (7), need (7), more (7), options (7), applications (7), properties (7), static (7), handle (7), during (7), how (7), stop (7), post (7), option (7), provisioning (7), only (7), migration (7), each (7), firewall (7), tcp (7), new (7), cli (7), enhanced (7), capabilities (7), commitments (7), startup (7), balancing (7), benchmark (7), manually (7), idle (7), deploy (7), back (7), mysql (7), zonal (7), remove (7), templates (7), backup (7), system (6), all (6), send (6), reserved (6), stacktype (6), nictype (6), include (6), body (6), part (6), two (6), make (6), preemption (6), rocky (6), names (6), same (6), steps (6), balanced (6), method (6), blocks (6), exfr (6), prod (6), bound (6), don (6), stack (6), stack_type (6), maximum (6), maxdistance (6), allow (6), drivers (6), between (6), get (6), required (6), identity (6), local (6), future (6), usage (6), security (6), data (6), updates (6), instant (6), files (6), nested (6), gpus (6), across (6), demand (6), pools (6), failover (6), extension (6), change (6), transfer (6), customized (6), understand (5), page (5), details (5), customize (5), needed (5), previously (5), omit (5), similar (5), either (5), version (5), specified (5), field (5), complete (5), googleapis (5), com (5), bulkinsert (5), deployment (5), model (5), first (5), europe (5), west4 (5), live (5), review (5), per (5), series (5), documentation (5), created (5), single (5), ranges (5), source (5), running (5), environment (5), calendar (5), clusters (5), monitoring (5), databases (5), accounts (5), errors (5), virtualization (5), nvidia (5), operations (5), node (5), highly (5), build (5), multi (5), modify (5), schedules (5), audit (5), symphony (5), email (5), scripts (5), manager (5), systems (5), tenancy (5), restore (5), control (5), workstation (5), faq (5), flex (5), português (4), español (4), sign (4), support (4), down (4), select (4), configures (4), chose (4), learn (4), termination_action (4), supports (4), gcp (4), gib (4), 192 (4), lssd (4), generates (4), such (4), https (4), global (4), reservation_block_name (4), placed (4), reservation_name (4), additional (4), section (4), max_distance (4), beta (4), collocated (4), subnets (4), range (4), attach (4), guides (4), usereadonly (4), account (4), template (4), services (4), apis (4), development (4), authenticate (4), provider (4), plan (4), tools (4), distributed (4), issues (4), ubuntu (4), bandwidth (4), pmu (4), ssd (4), share (4), upgrade (4), autoscaling (4), cross (4), requests (4), metrics (4), health (4), active (4), containers (4), container (4), database (4), postgresql (4), extensions (4), work (4), suspend (4), location (4), resize (4), graceful (4), locations (4), plans (4), replication (4), terms (3), content (3), apache (3), topology (3), sure (3), true (3), aren (3), reservation_bound (3), terminate (3), tab (3), not (3), before (3), instructions (3), distance (3), adjacent (3), collocation (3), close (3), latency (3), also (3), skip (3), settings (3), verify (3), check (3), sources (3), ip_range (3), profile (3), copy (3), private (3), predefined (3), read (3), assign (3), api (3), credentials (3), obtain (3), infrastructure (3), observability (3), analytics (3), automatic (3), troubleshooting (3), encryption (3), pro (3), serial (3), configurations (3), patterns (3), higher (3), discounts (3), cuds (3), manual (3), scalable (3), resilient (3), autoscale (3), balancer (3), prevent (3), testing (3), ibm (3), spectrum (3), perform (3), deploying (3), prepare (3), rhel (3), protect (3), certificates (3), query (3), state (3), disable (3), increase (3), schedule (3), lifecycle (3), design (3), non (3), pool (3), move (3), workstations (3), tpus (3), product (3), 한국어 (2), 日本語 (2), עברית (2), brasil (2), italiano (2), indonesia (2), français (2), américa (2), latina (2), deutsch (2), english (2), site (2), youtube (2), center (2), started (2), pricing (2), products (2), last (2), updated (2), 2026 (2), utc (2), licensed (2), under (2), feedback (2), next (2), submit (2), fields (2), networkinterfaces (2), accessconfigs (2), one_to_one_nat (2), natip (2), recommend (2), specifying (2), important (2), take (2), preempts (2), debian (2), namepattern (2), machinetype (2), initializeparams (2), disksizegb (2), disktype (2), sourceimage (2), read_write (2), scheduling (2), provisioningmodel (2), instanceterminationaction (2), reservationblocks (2), reservationaffinity (2), consumereservationtype (2), specific_reservation (2), key (2), values (2), anywhere (2), parameters (2), depend (2), corresponds (2), receive (2), ephemeral (2), valid (2), ipv4_ipv6 (2), whole (2), termination (2), affinity (2), but (2), applying (2), step (2), provide (2), click (2), outside (2), practice (2), traffic (2), regular (2), number_of_gvnic (2), num_gvnic (2), 65535 (2), udp (2), icmp (2), profiles (2), idpf (2), driver (2), internet (2), communication (2), configured (2), might (2), these (2), labels (2), tags (2), legacy (2), useexternalip (2), chosen (2), subnetworks (2), admin (2), iam (2), idp (2), federated (2), after (2), process (2), preempted (2), large (2), sdk (2), languages (2), frameworks (2), costs (2), industry (2), solutions (2), hybrid (2), multicloud (2), pipelines (2), hosting (2), quota (2), commitment (2), standard (2), full (2), registration (2), sles (2), export (2), port (2), collecting (2), nodes (2), report (2), arm (2), core (2), rdp (2), general (2), app (2), analyze (2), visible (2), cores (2), automatically (2), committed (2), generation (2), underutilized (2), insights (2), utilization (2), globally (2), web (2), autohealing (2), iis (2), shared (2), reserve (2), autoscaler (2), autoscalers (2), activity (2), kubernetes (2), directory (2), setting (2), redis (2), cloning (2), writer (2), always (2), alwayson (2), file (2), volumes (2), synchronization (2), switch (2), byol (2), payg (2), byos (2), changes (2), families (2), guide (2), securing (2), event (2), simulate (2), together (2), repairs (2), repair (2), failures (2), suspended (2), stopped (2), target (2), distribution (2), physical (2), cancel (2), once (2), limit (2), decrease (2), basic (2), consistent (2), scoped (2), appliances (2), asynchronous (2), throughput (2), encrypt (2), format (2), mount (2), memory (2), extreme (2), organization (2), user (2), through (2), advanced (2), preemptible (2), models (2), rtx (2), vws (2), constraints (2), accelerator (2), reference (2), technology (2), areas (2), subscribe, newsletter, our, third, decade, climate, join, cookies, privacy, tech, twitter, blog, engage, training, certification, architecture, getting, github, status, release, notes, community, forums, contact, sales, marketplace, easy, easytounderstand, solved, problem, solvedmyproblem, otherup, hard, hardtounderstand, incorrect, sample, incorrectinformationorsamplecode, missing, missingtheinformationsamplesineed, otherdown, tell, except, otherwise, noted, java, registered, trademark, oracle, its, affiliates, developers, creative, commons, attribution, mpi, what, onhostmaintenance, automaticrestart, both, limitations, describe, groupplacementpolicy, insert, max, makes, effort, attempts, possible, sensitive, requires, compactness, lower, ensures, closer, increases, chance, some, won, preview, tip, search, previous, successfully, central1, rather, than, because, allows, including, correspond, bin, bash, seq, done, rule, ping, offset, avoid, overlap, shell, window, there, nine, separate, creates, 8896, transmission, unit, mtu, configuring, auto, instruction, follow, documented, provided, eight, total, vnics, normal, gve, least, includes, able, write, setserviceaccount, setlabels, settags, setmetadata, instancetemplates, contain, exact, expand, granting, folders, organizations, networkadmin, instanceadmin, ask, administrator, grant, haven, already, verifies, selecting, installed, latest, components, note, init, installation, initialize, differs, compactly, they, eligible, managing, begin, ways, tightly, coupled, document, explains, computing, identical, independent, save, categorize, preferences, stay, organized, collections, home, concurrent, operation, renewal, nvme, resizing, common, pay, viewing, output, diagnostic, sudoers, screenshots, generate, bug, blackwell, soft, lockups, bus, locks, rescue, inaccessible, dumps, kernel, panic, fstab, unresponsive, suspension, reboots, shutdowns, connectivity, tips, nics, dpdk, improve, resiliency, reduce, threads, sustained, merge, split, extend, renew, without, purchase, cost, savings, dynamic, overcommit, cpus, floating, reliable, backends, routing, scaling, consuming, consume, combine, cud, sharing, decisions, signals, serving, predictions, organize, replica, states, observe, reports, tensorflow, inference, tensorrt5, learning, monte, carlo, methods, spark, forwarding, others, integrate, clustering, exchange, 2016, sharepoint, strategies, passive, inactive, setups, built, integrations, validation, were, deployed, transition, test, hammerdb, bucket, aws, ec2, pacemaker, s2d, netapp, hot, standby, drbd, architectures, client, mailjet, mailgun, sendgrid, sending, blue, green, deployments, lamp, joomla, asp, interactive, mongodb, flask, terraform, servers, over, mtls, random, generator, virtio, rng, accurate, protocol, ntp, append, els, restrictions, packages, deprecate, versions, trusted, red, hat, knowledgebase, functionality, risks, controls, confidential, kek, secure, expiration, shielded, attributes, notices, faulty, unmanaged, affect, preserved, turn, off, alternate, fails, repairing, maintain, upgrades, override, selectively, accelerate, out, outage, rebalance, reenable, proactive, redistribution, shape, distribute, define, info, accidental, deletion, spread, edit, rename, reset, resume, restart, referrers, uuid, recover, corrupted, duplicate, clones, alerts, backups, vss, regionally, different, protection, considerations, synchronous, consistency, failback, replicate, provisioned, iops, limits, evaluate, help, kms, customer, supplied, symbolic, links, device, ram, exapools, ptr, record, failure, rates, restrict, automate, password, require, powershell, sac, securely, apps, root, vpn, bastion, iap, connection, browser, connections, within, iso, imported, prerequisites, importing, exporting, bring, own, path, detach, reattach, constraint, accounting, provision, examples, slurm, team, slice, slices, adds, scenarios, deterministic, base, requirements, ready, aci, historical, alternative, displays, ops, logging, public, has, gpudirect, enabled, ultra, minimum, platform, hostname, quickstarts, office, licensing, premium, strategy, gemini, bare, metal, accelerators, platforms, scores, purpose, discover, free, main,
Text of the page (random words):
e configuring only one gvnic network interface you can use the default vpc network and the auto subnet that s in the same region as the instance to create a network for the irdma network interface see create a vpc network with a falcon vpc network profile use the default value for the maximum transmission unit mtu for a falcon vpc network which is 8896 script you can create up to nine gvnic network interfaces and one irdma network interface per instance each network interface must attach to a separate network to create the networks you can use the following script which creates two networks for gvnic and one network for irdma optional before running the script list the falcon vpc network profiles to verify there is one available gcloud compute network profiles list copy the following code and run it in a linux shell window bin bash set the number of gvnic interfaces to create you can create up to 9 num_gvnic number_of_gvnic create regular vpc networks and subnets for the gvnic interfaces for n in seq 0 num_gvnic 1 do gcloud compute networks create gvnic_name_prefix net n subnet mode custom gcloud compute networks subnets create gvnic_name_prefix sub n network gvnic_name_prefix net n region region range 10 n 0 0 16 gcloud compute firewall rules create gvnic_name_prefix internal n network gvnic_name_prefix net n action allow rules tcp 0 65535 udp 0 65535 icmp source ranges 10 0 0 0 8 done create ssh firewall rules gcloud compute firewall rules create gvnic_name_prefix ssh network gvnic_name_prefix net 0 action allow rules tcp 22 source ranges ip_range optional create a firewall rule for the external ip address for the first gvnic network interface gcloud compute firewall rules create gvnic_name_prefix allow ping net 0 network gvnic_name_prefix net 0 action allow rules icmp source ranges ip_range create a falcon vpc network for the cloud rdma network interface gcloud compute networks create rdma_name_prefix irdma network profile zone vpc falcon subnet mode custom create a subnet in the falcon vpc network gcloud compute networks subnets create rdma_name_prefix irdma sub network rdma_name_prefix irdma region region range 10 2 0 0 16 offset to avoid overlap with gvnic subnet ranges replace the following number_of_gvnic the number of gvnic interfaces to create specify a number from 1 to 9 gvnic_name_prefix the name prefix to use for the regular vpc network and subnet that uses a gvnic nic type region the region where you want to create the networks this must correspond to the zone specified for the network profile flag when creating the falcon vpc network for example if you specify the zone as europe west4 b then your region is europe west4 ip_range the range of ip addresses outside of the vpc network to use for the ssh firewall rules as a best practice specify the specific ip address ranges that you need to allow access from rather than all ipv4 or ipv6 sources don t use 0 0 0 0 0 or 0 as a source range because this allows traffic from all ipv4 or ipv6 sources including sources outside of google cloud rdma_name_prefix the name prefix to use for the vpc network and subnet that uses the irdma nic type zone the zone where you want to create the networks and compute instances use either us central1 a or europe west4 b optional to verify that the vpc network resources are created successfully check the network settings in the google cloud console in the google cloud console go to the vpc networks page go to vpc networks search the list for the networks that you created in the previous step to view the subnets firewall rules and other network settings click the name of the network optional create a placement policy tip to place your vms in single or adjacent blocks specify vm placement by creating and applying a placement policy to your vms to place vms on a specific block skip this step and provide the name of the block when you create vms you can specify vm placement by creating a compact placement policy when you apply a compact placement policy to your vms compute engine makes best effort attempts to create vms that are as close to each other as possible if your application is latency sensitive and requires maximum compactness then specify the maxdistance field preview when you create a compact placement policy a lower maxdistance value ensures closer vm placement but it also increases the chance that some vms won t be created to create a compact placement policy select one of the following options gcloud to create a compact placement policy use the gcloud beta compute resource policies create group placement command gcloud beta compute resource policies create group placement policy_name collocation collocated max distance max_distance region region replace the following policy_name the name of the compact placement policy max_distance the maximum distance configuration for your vms the value must be 3 to place vms in the adjacent blocks or 2 to place vms in the same block for information about the maximum number of vms supported for each maxdistance per machine series see about compact placement policies in the compute engine documentation region the region where you want to create the compact placement policy specify a region in which the machine type that you want to use is available for information about regions see available regions and zones rest to create a compact placement policy make a post request to the beta resourcepolicies insert method in the request body include the collocation field set to collocated and the maxdistance field post https compute googleapis com compute beta projects project_id regions region resourcepolicies name policy_name groupplacementpolicy collocation collocated maxdistance max_distance replace the following project_id your project id policy_name the name of the compact placement policy max_distance the maximum distance configuration for your vms the value must be 3 to place vms in the adjacent blocks or 2 to place vms in the same block for information about the maximum number of vms supported for each maxdistance per machine series see about compact placement policies in the compute engine documentation region the region where you want to create the compact placement policy specify a region in which the machine type that you want to use is available for information about regions see available regions and zones create vm instances in bulk the instructions in this section describe how to create h4d vms in bulk review the following limitations before creating h4d instances with cloud rdma you can t use live migration during host maintenance events with instances that have a cloud rdma network interface you must configure the instance to terminate during maintenance events the gvnic network interface can have only one ipv6 address either internal or external but not both you can use only ipv4 addresses with irdma network interfaces and falcon vpc networks gcloud to create vms in bulk use the gcloud compute instances create command the parameters that you need to specify depend on the consumption option that you are using for this deployment select the tab that corresponds to your consumption option s provisioning model reservation bound start with the following gcloud compute instances create command gcloud compute instances bulk create name pattern name_pattern count count machine type machine_type image family image_family image project image_project instance termination action delete maintenance policy terminate region region boot disk type hyperdisk balanced boot disk size disk_size complete the following steps replace the following name_pattern the name pattern for the instances for example using vm for the name pattern generates instances with names such as vm 1 and vm 2 up to the number specified by the count flag count the number of instances to create machine_type the machine type to use for the instances use one of the h4d machine types for example h4d highmem 192 lssd image_family the image family of the os image that you want to use for example rocky linux 9 optimized gcp for a list of supported os images see supported operating system choose an os image version that supports the irdma interface image_project the project id for the os image for example rocky linux cloud region specify a region in which the machine type that you want to use is available for example europe west4 for information about available regions see available regions and zones disk_size optional the size of the boot disk in gib the value must be a whole number optional if you chose to use a compact placement policy include the resource policies flag resource policies policy_name replace policy_name with the name of the compact placement policy to specify the reservation do one of the following if you are using a placement policy or if vms can be placed anywhere in your reservation block then add the following flags to the command provisioning model reservation_bound reservation affinity specific reservation reservation_name replace reservation_name with the name of the reservation for example h4d highmem exfr prod if you aren t using a compact placement policy and you want the instances placed in a specific block then add the following flags to the command provisioning model reservation_bound reservation affinity specific reservation reservation_block_name replace reservation_block_name with the name of a block in the reservation for example h4d highmem exfr prod reservationblocks h4d highmem exfr prod block 1 to view the reservation name or the available reservation blocks see view capacity optional to configure the instances to use cloud rdma add the flags similar to the following to the command this example configures two gvnic network interfaces and one irdma network interface network interface nic type gvnic network gvnic_name_prefix net 0 subnet gvnic_name_prefix sub 0 stack type stack_type address external_ipv4_address network interface nic type gvnic network gvnic_name_prefix net 1 subnet gvnic_name_prefix sub 1 no address network interface nic type irdma network rdma_name_prefix irdma subnet rdma_name_prefix irdma sub stack type ipv4_only no address replace the following gvnic_name_prefix the name prefix you used when creating the vpc network and subnet for the gvnic interface for the first gvnic network interface you can omit the network and subnet flags to use the default network instead stack_type optional the stack type for the gvnic network interface stack_type must be one of ipv4_only or ipv4_ipv6 the default value is ipv4_only external_ipv4_address optional a static external ipv4 address to use with the network interface you must have previously reserved an external ipv4 address do one of the following specify a valid ipv4 address from the subnet use the flag no address instead if you don t want the network interface to have an external ip address specify address if you want the interface to receive an ephemeral external ip address to specify an external ipv6 address use the flag external ipv6 address instead rdma_name_prefix the name prefix that you used when creating the vpc network and subnet for the irdma interface optional add additional flags to customize the rest of the instance properties as needed run the command spot start with the following gcloud compute instances create command gcloud compute instances bulk create name pattern name_pattern count count machine type machine_type image family image_family image project image_project region region boot disk type hyperdisk balanced boot disk size disk_size provisioning model spot instance termination action termination_action complete the following steps replace the following name_pattern the name pattern for the instances for example using vm for the name pattern generates instances with names such as vm 1 and vm 2 up to the number specified by the count flag count the number of instances to create machine_type the machine type to use for the instances use one of the h4d machine types for example h4d highmem 192 lssd image_family the image family of the os image that you want to use for example rocky linux 9 optimized gcp for a list of supported os images see supported operating system choose an os image version that supports the irdma interface image_project the project id for the os image for example rocky linux cloud region specify a region in which the machine type that you want to use is available for example europe west4 for information about available regions see available regions and zones disk_size optional the size of the boot disk in gib the value must be a whole number termination_action the action to take when compute engine preempts the instance either stop default or delete important make sure your application can handle preemption for example we recommend that you handle preemption by specifying a shutdown script during instance creation learn how to handle preemption with a shutdown script optional if you chose to use a compact placement policy then add the following flag to the command resource policies policy_name replace policy_name with the name of the compact placement policy optional to configure the instances to use cloud rdma add the flags similar to the following to the command this example configures two gvnic network interfaces and one irdma network interface network interface nic type gvnic network gvnic_name_prefix net 0 subnet gvnic_name_prefix sub 0 stack type stack_type address external_ipv4_address network interface nic type gvnic network gvnic_name_prefix net 1 subnet gvnic_name_prefix sub 1 no address network interface nic type irdma network rdma_name_prefix irdma subnet rdma_name_prefix irdma sub stack type ipv4_only no address replace the following gvnic_name_prefix the name prefix you used when creating the vpc network and subnet for the gvnic interface for the first gvnic network interface you can omit the network and subnet flags to use the default network instead stack_type optional the stack type for the gvnic network interface stack_type must be one of ipv4_only or ipv4_ipv6 the default value is ipv4_only external_ipv4_address optional a static external ipv4 address to use with the network interface you must have previously reserved an external ipv4 address do one of the following specify a valid ipv4 address from the subnet use the flag no address instead if you don t want the network interface to have an external ip address specify address if you want the interface to receive an ephemeral external ip address to specify an external ipv6 address use the flag external ipv6 address instead rdma_name_prefix the name prefix that you used when creating the vpc network and subnet for the irdma interface optional add additional flags to customize the rest of the instance properties as needed run the command rest to create vm instances in bulk make a post request to the instances bulkinsert method the parameters that you need to specify depend on the consumption option that you are using for...
|
