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):
ness 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 this deployment select the tab that corresponds to your consumption option s provisioning model reservation bound start with the following post request to the instances bulkinsert method post https compute googleapis com compute v1 projects project_id zones zone instances bulkinsert namepattern name_pattern count count instanceproperties machinetype machine_type disks boot true initializeparams disksizegb disk_size disktype hyperdisk balanced sourceimage projects image_project global images family image_family mode read_write type persistent scheduling provisioningmodel reservation_bound instanceterminationaction delete onhostmaintenance terminate automaticrestart true complete the following steps replace the following project_id the project id of the project where you want to create the instances zone specify a zone in which the machine type that you want to use is available if you are using a compact placement policy then use a zone in the same region as the compact placement policy for information about the regions where h4d machine types are available see available regions and zones 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 field 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 disk_size the size of the boot disk in gib image_project the project id for the os image for example debian cloud 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 optional if you chose to use a compact placement policy include the resourcepolicies parameter in the request body as part of the instanceproperties parameter resourcepolicies projects project_id regions region resourcepolicies 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 to the request body as part of the instanceproperties parameter reservationaffinity consumereservationtype specific_reservation key compute googleapis com reservation name values 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 or you want the instances placed in a specific block then add the following to the request body as part of the instanceproperties parameter reservationaffinity consumereservationtype specific_reservation key compute googleapis com reservation name values 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 if you want to configure the instances to use cloud rdma then include a parameter block similar to the following to the request body as part of the instanceproperties parameter this example configures two gvnic network interfaces and one irdma network interface networkinterfaces network gvnic_name_prefix net 0 subnetwork gvnic_name_prefix sub 0 accessconfigs type one_to_one_nat name external ip natip external_ipv4_address stacktype ipv4_only nictype gvnic network gvnic_name_prefix net 1 subnetwork gvnic_name_prefix sub 1 stacktype ipv4_only nictype gvnic network rdma_name_prefix irdma subnetwork rdma_name_prefix irdma sub stacktype ipv4_only nictype irdma replace the following gvnic_name_prefix the name prefix that you used when creating the vpc network and subnet for the gvnic interface for the gvnic network interface you can omit the network and subnetwork fields to use the default network instead external_ipv4_address optional a static external ipv4 address to use with the network interface you must have previously reserved an external ipv4 ad...
|
