iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
In 2026, the data center supply chain is constrained less by servers than by coordinated delivery of power, grid connections, transformers, cooling, specialized labor, networking, and high-density IT hardware. A data center is not commercially usable when GPUs arrive alone; the facility must also be energized, cooled, connected, commissioned, and ready for customers.
AI has changed the constraint from a hardware-allocation problem into a synchronized infrastructure problem. Higher rack densities are increasing electrical and thermal requirements while hyperscalers, cloud providers, colocation operators, utilities, and construction firms compete for many of the same scarce inputs.
Key takeaways
- JLL reported average 2026 data-center equipment lead times of 33 weeks globally and 42 weeks in the United States, with the U.S. figure 83% above 2019 levels.
- According to the IEA, data-center electricity demand grew 17% in 2025, while AI-server power density increased elevenfold between 2020 and 2025.
- Large power transformers can require up to four years to procure, according to an IEA transmission-industry survey, making grid equipment a structural constraint rather than an ordinary purchasing delay.
- JLL reported that 57% of projects in its 2025 sample experienced construction delays of at least three months.
- Power secured, equipment delivered, and building complete are separate milestones; none alone proves that a facility can enter revenue service.
- The most credible projects show an executable chain from utility energization and long-lead equipment through cooling validation, network commissioning, customer acceptance, and operations staffing.
What does the data center supply chain include?
The data center supply chain includes every input required to turn a site and a capacity forecast into reliable, revenue-producing compute. The chain starts with demand planning and land, then crosses utilities, electrical equipment, generators, cooling, construction, semiconductors, servers, networking, software, logistics, labor, commissioning, and maintenance.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Scan for outdated or missing drivers - takes under a minuteDriver Scan →A useful end-to-end model is:
AI demand → chips and memory → servers and networking → racks → power distribution → cooling and heat rejection → building and site → grid interconnection or generation → commissioning → usable capacity.
#1 Best Overall
| Supply-chain layer | What it contains | Typical failure if missed |
|---|---|---|
| Demand and planning | Workload forecasts, tenants, accelerator generations, capacity phases, financing | Capacity is approved before its technical or commercial assumptions are fixed |
| Site and utility | Land, permits, transmission, distribution, interconnection studies, fiber and water | A site is announced but cannot receive its promised power on schedule |
| Electrical infrastructure | Transformers, switchgear, UPS systems, batteries, busway, PDUs, protection and controls | A finished building remains unenergized or cannot support its planned load |
| Thermal infrastructure | Chillers, cooling towers, CDUs, cold plates, pumps, heat exchangers, leak detection | Racks are installed but cannot operate at their design density |
| IT hardware | Accelerators, CPUs, HBM, storage, hosts, racks, firmware and software | Compute is allocated but the complete validated system is incomplete |
| Networking | Switches, optics, fiber, cables, fabric management, external connectivity | A powered cluster cannot deliver usable distributed computing |
| Construction and logistics | Steel, concrete, modular rooms, cable tray, fire systems, transport and cranes | Equipment arrives before the site is ready, or site work waits for a module |
| People and operations | Electricians, controls engineers, commissioning agents, network specialists and operators | Physical completion is followed by testing, staffing or acceptance delays |
Why has AI made the data center supply chain harder?
AI has made the data center supply chain harder because AI clusters combine higher rack density, larger simultaneous deployments, more demanding thermal systems, heavier networking requirements, faster refresh cycles, and greater dependence on advanced packaging and high-bandwidth memory.
According to the IEA executive summary on energy and AI, data-center electricity demand grew 17% in 2025. The same IEA source says the largest technology companies spent more than $400 billion on capital expenditure in 2025 and expects that spending to increase by 75% in 2026. Those figures describe demand pressure, not guaranteed completed capacity.
The IEA also estimates that AI-server power density increased elevenfold between 2020 and 2025 and could increase another fourfold by 2027. The figures apply to AI-server power density, not to every server or every data center. The practical consequence is that a room designed around conventional air-cooled racks may not support the electrical load, heat rejection, floor layout, or service procedures required by a later AI deployment.
The Tool Desk
Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →AI infrastructure also has a system-level dependency problem. An accelerator shipment is not installed compute without matching HBM, host servers, network fabric, optics, racks, liquid-cooling equipment, power distribution, firmware, and commissioning. A project can therefore have GPUs available while remaining unable to deliver a working cluster.
Which data center supply-chain constraints matter most?
Power availability and interconnection are currently the highest-impact constraints, followed by grid equipment, high-density cooling, complete AI systems, and commissioning capacity. The ranking changes by project, but the order reflects how one missing upstream dependency can hold back an entire facility.
1. Power availability and grid interconnection
Power is increasingly the first site-selection criterion because a data hall without dependable electricity cannot generate revenue regardless of its building quality or server allocation. JLL identifies speed to power as the primary criterion driving site selection, ahead of community support, latency, and customer proximity, in its 2026 data center outlook.
“Power secured” must be examined carefully. A credible project should distinguish a utility study, an interconnection request, a conditional capacity reservation, an executed agreement, a construction milestone, and a firm energization date. A utility commitment may depend on a future substation, transmission upgrade, protection study, or customer-owned transformer that has not yet been delivered.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
2. Transformers, switchgear, and electrical distribution
Transformers and medium-voltage equipment frequently determine whether a nearly complete building can be energized. Relevant equipment includes utility-side and customer-owned transformers, medium-voltage switchgear, low-voltage switchboards, busway, automatic transfer switches, generator paralleling gear, protection systems, UPS equipment, batteries, rack PDUs, grounding, and bonding.
According to the IEA report on building the future transmission grid, cable procurement can take two to three years and large power transformers can take up to four years. The IEA figures come from the transmission industry and refer to those equipment categories broadly; they do not mean every data-center transformer takes four years.
Grid equipment is especially difficult because data centers compete with transmission expansion, industrial electrification, renewable projects, and other large loads. A project can have its shell completed and its servers purchased while waiting for one transformer or a protection-and-control package.
3. Generators, batteries, and on-site power
On-site generation can reduce dependence on a slow grid interconnection, but on-site generation does not make grid power irrelevant. A project still needs fuel logistics, protection studies, backup arrangements, maintenance, regulatory approval, emissions compliance, and a plan for operating when the grid is unavailable.
Possible approaches include gas turbines, reciprocating engines, fuel cells, battery energy storage, renewable generation paired with storage, temporary generation, microgrids, heat recovery, and future nuclear or advanced-reactor concepts. Vertiv describes bring-your-own-power and cooling solutions that combine on-site generation with integrated infrastructure for sites facing slow grid interconnection. That is a vendor solution description, not independent evidence that every site can deploy the model quickly.
Rank #2
| Power strategy | Potential advantage | New exposure |
|---|---|---|
| Grid power | Lower operating complexity and an established utility relationship | Interconnection queues, transmission upgrades and grid congestion |
| On-site generation | Can shorten time to initial power at constrained sites | Fuel, emissions, noise, permitting, maintenance and stranded-asset risk |
| Battery storage | Supports bridging, peak management and microgrid operation | Finite duration, replacement requirements, fire protection and capital cost |
| Hybrid microgrid | Combines grid, generation and storage for flexibility | More controls, protection, integration and operating complexity |
4. High-density cooling
High-density AI deployments are pushing traditional room-air cooling toward its limits. CBRE reports that GPU-intensive workloads are driving adoption of direct-to-chip and immersion liquid cooling in its North America data center trends report.
Liquid cooling is not one architecture. It may mean direct-to-chip cooling using facility water loops, single-phase immersion, two-phase immersion, rear-door heat exchangers, or a hybrid air-liquid design. A selected architecture affects cold plates, manifolds, hoses, coolant distribution units, pumps, heat exchangers, water chemistry, leak detection, floor loading, maintenance access, and heat rejection.
Switching from air cooling to liquid cooling late in a project can force changes to mechanical rooms, pipe routes, rack layouts, electrical loads, floor structures, fire protection, and operating procedures. Liquid cooling is highly suitable for some dense AI racks, but liquid cooling is not automatically necessary for every workload, server, climate, retrofit, or facility.
Recommended Free Tools
Vendor product portfolios illustrate the breadth of the category. Vertiv’s product portfolio includes critical power, modular infrastructure, and liquid-cooling products such as cooling distribution units. Schneider Electric’s prefabricated module information describes IT pods supporting more than 40 high-density racks in configurations that depend on the selected design.
5. Accelerators, memory, and advanced packaging
AI hardware demand reaches beyond GPUs. The relevant components include other accelerators, HBM, advanced packaging, high-speed networking silicon, optical transceivers, CPUs, host systems, storage, power-management components, and complete server platforms.
The IEA identifies simultaneous pressure across electricity, grid connections, manufacturing capacity, chips, power electronics, and transformers. The practical procurement question is not whether one accelerator model is available. The question is whether a validated platform—including memory, hosts, networking, power, cooling, software, and support—can be delivered within the project’s commissioning window.
6. Networking and interconnect
AI clusters depend on high-bandwidth fabric between accelerators, spine-leaf switching, optical modules, fiber, high-speed copper or active electrical cables, storage networking, timing, management, and external connectivity.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Scan for outdated or missing drivers - takes under a minute3Repair Windows errors before they cause bigger problemsNetworking becomes a last-mile bottleneck when racks and power are ready but the internal fabric or external connection is not commissioned. Procurement teams should verify port counts, optics compatibility, cable routing, bend radius, fiber availability, network power and cooling, firmware, interoperability testing, and vendor support. Theoretical rack capacity is not the same as usable cluster capacity.
7. Construction materials and prefabrication
Construction dependencies include steel, concrete, structural modules, electrical rooms, mechanical skids, racks, containment, fire-rated assemblies, cable tray, roofing, fire suppression, security systems, and control systems.
According to JLL’s 2026 data center outlook, average global shell-and-core construction cost rose from $7.7 million per MW in 2020 to $10.7 million per MW in 2025, with a projected 2026 average of $11.3 million per MW. JLL says the figures exclude tenant fit-out and that AI tenant technology fit-out can cost up to $25 million per MW. Those are different cost layers and should not be combined into a single shell-construction price.
Prefabrication can compress schedules by moving work into factories while site preparation proceeds in parallel. Schneider describes modular products combining power, cooling, racks, batteries, controls, and management systems in factory-assembled units. Prefabrication is schedule compression, not a guarantee of faster delivery. Factory backlog, transportation permits, cranes, foundations, local code review, and commissioning labor can become the new bottlenecks.
8. Skilled labor and commissioning
Specialized labor is a supply-chain input, not merely a construction overhead. Projects need electricians, high-voltage technicians, controls engineers, mechanical engineers, welders, pipefitters, commissioning agents, network engineers, liquid-cooling specialists, construction managers, utility-interconnection experts, reliability staff, and operations personnel.
Rank #3
More than half of respondents reported difficulty finding qualified candidates in the Uptime Institute Global Data Center Survey 2026. Staffing shortages can delay testing, maintenance readiness, and operational acceptance even when equipment is physically present.
Commissioning is a separate phase with its own risks. Protection-relay configuration, generator synchronization, UPS and battery testing, cooling-water balancing, leak testing, network interoperability, fire approvals, load-bank testing, failure-mode testing, and customer acceptance testing can all delay revenue service.
Are data center supply chains improving in 2026?
Data center supply chains are improving unevenly: some equipment lead times have stabilized year over year, but lead times remain well above pre-2020 levels and grid equipment remains structurally constrained.
JLL reports a 33-week global average equipment lead time and a 42-week U.S. average in 2026. JLL also reports that developers are preordering selected materials as much as 24 months ahead, while operators and developers of scale are holding six to twelve months of strategic inventory. Stabilization means conditions are no longer worsening as quickly in every category; stabilization at an elevated level is not normalization.
Demand growth can absorb newly added manufacturing capacity. A supplier may increase output while the buyer’s required quantity increases faster. Grid equipment has an even longer cycle because manufacturing expansion, engineering qualification, factory testing, transport, utility studies, and installation all take time.
| Indicator | 2026 figure or outlook | How to interpret it |
|---|---|---|
| Global equipment lead time | 33 weeks | Average reported by JLL; not a promise for a specific component |
| U.S. equipment lead time | 42 weeks | JLL says this is 83% above 2019 levels |
| Projects delayed at least three months | 57% in 2025 | Applies to JLL’s project sample and methodology |
| Selected-material advance procurement | Up to 24 months | Applies to critical selected materials, not every purchase |
| Strategic inventory | Six to twelve months | Reported for operators and developers of scale |
| Large transformer procurement | Up to four years | IEA transmission-industry survey figure |
Why are data centers moving to secondary markets?
Data centers are moving to secondary markets primarily because power availability, land, permitting, construction capacity, and fiber can matter more than proximity to an established hub. CBRE reports that power and grid constraints are affecting development timelines and site selection in major hubs including Northern Virginia, Chicago, London, and Frankfurt in its global data center trends report.
A secondary market can offer available power and land while lacking fiber density, specialized labor, supplier ecosystems, water, or nearby customers. Long-distance networking can reduce the penalty for geographic distance, but it cannot eliminate latency, regulatory, resilience, or connectivity requirements.
Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallSite selection should therefore score power, energization date, land, permitting, fiber, labor, water, heat rejection, logistics, customer latency, community acceptance, and future expansion together. “Power available” is not enough if the site cannot obtain network connectivity or qualified commissioning staff.
Which supply-chain strategies reduce schedule risk?
The strongest strategy is coordinated early procurement tied to a frozen reference design, explicit alternate suppliers, phased capacity, and independent verification of utility and factory milestones.
Early procurement and deposits
Long-term orders and deposits can reserve production slots for transformers, switchgear, UPS systems, generators, cooling equipment, network hardware, and modules. Early procurement reduces queue risk but increases exposure to design changes, cancellation costs, storage, warranty periods, and technology obsolescence.
Strategic inventory
Inventory should be based on criticality, failure probability, replacement lead time, standardization, shelf life, storage cost, compatibility, and future support. Holding GPUs is different from holding switchgear or spare transformers: compute hardware can become obsolete quickly, while heavy electrical equipment is difficult and expensive to store or redeploy.
Free tools Windows power users keep installed
One-click scans. No signup required.
Standardization and multiple approved suppliers
Standard rack layouts, electrical blocks, cooling interfaces, controls, and commissioning procedures make phased expansion more repeatable. A multiple-supplier design increases flexibility, but a second supplier is not real redundancy if both suppliers depend on the same sub-tier transformer maker, semiconductor source, casting facility, port, or scarce labor pool.
Modular and prefabricated construction
Modules can allow factory testing, parallel site work, repeatable expansion, and reduced site labor. The trade-off is reduced flexibility after design freeze and greater exposure to factory capacity, transport, crane access, local code requirements, and integration between modules and site systems.
Liquid cooling
Liquid cooling can support dense AI deployments, but the selection must include the IT manufacturer, facility water loop, coolant or water chemistry, leak detection, heat rejection, service procedures, spare parts, and technician training. A mixed air-liquid architecture may be more practical when only a portion of the facility contains high-density AI racks.
On-site power
On-site power can reduce initial interconnection dependence, but the project must model fuel supply, emissions, permitting, noise, maintenance, backup, controls, and the possibility that grid power arrives earlier or later than expected. A temporary bridge can become a stranded asset if the final operating model changes.
Retrofitting existing facilities
Retrofitting can avoid some greenfield construction and may provide faster access to a building, electrical service, or network connection. Retrofit feasibility depends on floor loading, pipe routes, heat rejection, electrical capacity, ceiling height, fire protection, water systems, and the ability to isolate construction from live operations.
What is the difference between announced capacity and usable capacity?
Announced capacity is a public development claim; usable capacity is power-available, cooled, connected, commissioned capacity that can support a customer or production workload. The distinction is essential for investors, tenants, operators, and policymakers evaluating supply.
| Milestone | What it proves | What it does not prove |
|---|---|---|
| Announcement | An organization has disclosed an intended project | Land, financing, power, equipment, or tenants are secured |
| Land secured | The project controls a site | The site can receive power, fiber, water, permits, or equipment |
| Interconnection requested | A utility process has begun | A firm energization date or deliverable capacity |
| Power contract executed | Commercial utility arrangements exist | Customer-owned transformers, substations, or upgrades are complete |
| Construction started | Physical site work has begun | Long-lead equipment or IT fit-out is available |
| Equipment ordered | A purchase commitment exists | Factory acceptance, shipment, installation, or interoperability |
| Shell completed | The building envelope and core are substantially complete | Energization, cooling validation, networking, or tenant readiness |
| Energization achieved | Power can reach the facility or a defined phase | Stable full-load operation or customer-ready compute |
| IT fit-out completed | Servers, racks, networking, and supporting systems are installed | Successful testing, acceptance, staffing, or production service |
| Revenue service | Capacity is operational and commercially usable | Future phases or expansion capacity |
What can still go wrong after power and equipment are secured?
Power is available but the project misses schedule
The interconnection may depend on a future substation, the approved capacity may be smaller than the announced build, customer-owned transformers may be late, protection studies may be incomplete, or temporary generation may not support the final load.
Equipment has shipped but the data hall is not operational
Controls integration, cooling imbalance, network commissioning, fire-safety approval, generator synchronization, firmware compatibility, load-bank testing, or telemetry can still block operation.
Free tools Windows power users keep installed
One-click scans. No signup required.
Liquid cooling is installed but performance is inadequate
Actual rack heat may exceed the design assumption, flow or water temperature may be incorrect, cold plates may not match the server, CDU capacity may be insufficient, heat rejection may be undersized, or maintenance staff may lack leak-response training.
Two suppliers provide no real redundancy
Both suppliers may use the same sub-tier transformer manufacturer, semiconductor source, specialized casting supplier, shipping route, or labor market. Supplier count must be measured at critical sub-tier dependencies, not only at the contract level.
Modular construction is not faster
A modular project can still wait for factory production, design approval, transport permits, site foundations, utility connection, local code review, or commissioning personnel.
Demand changes after procurement
AI utilization may be lower than projected, accelerator generations may change faster than procurement cycles, customers may defer capacity, financing costs may rise, or regulation may slow construction. Over-ordering long-lived electrical equipment creates a different risk from over-ordering rapidly obsolete compute hardware.
The Tool Desk
Outbyte PC Repair FREERepair Windows errors before they cause bigger problemsFix Now →Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →How should investors evaluate whether a data center is executable?
Investors and lenders should test the evidence behind every major milestone rather than rely on the project’s announced megawatt figure.
Best Value
- Is “power secured” supported by a legally documented agreement?
- Does the utility provide a firm energization date for each planned phase?
- Are transmission upgrades, substations, transformers, and switchgear ordered?
- Are factory production slots and factory-acceptance-test dates documented?
- Is the tenant committed, and is its AI fit-out funded separately from shell and core?
- Are accelerators allocated, or is the business case based on an assumption about future availability?
- Does the design support the target rack density and cooling architecture?
- Can the project operate if AI demand, accelerator economics, or financing conditions change?
- Does the site have fuel, water, fiber, local labor, spares, and commissioning capacity?
- Are milestones measured through energization, IT fit-out, acceptance testing, and revenue service rather than only construction completion?
What should procurement teams verify before placing orders?
Procurement teams should prioritize schedule criticality and replacement difficulty, not purchase price alone. A practical sequence begins with transformers and medium-voltage gear, followed by switchgear and protection systems, UPS and batteries, generators and fuel systems, cooling equipment and CDUs, network switches and optics, racks and busway, accelerators and servers, controls, and replacement parts.
Every critical supplier review should verify a confirmed production slot, factory-acceptance-test schedule, purchase-order status, qualified alternates, shared sub-tier dependencies, local service coverage, spare-parts availability, accelerator compatibility, liquid-cooling readiness, controls and DCIM interoperability, warranty terms, cybersecurity of monitoring systems, utility certainty, fuel and water availability, commissioning capacity, and phase-expansion capability.
Procurement should also classify each item as owned inventory, reserved capacity, allocated production, or a verbal estimate. Those are not equivalent forms of supply assurance.
Which infrastructure vendors fit the current supply-chain problem?
Enterprise data-center infrastructure is generally engineered to order and quote-based, so vendor selection should focus on delivery evidence, compatibility, service coverage, spares, and integration scope rather than public retail pricing.
| Vendor | Strongest fit | Relevant capabilities | Main caution |
|---|---|---|---|
| Schneider Electric | Large projects seeking an integrated ecosystem | Prefabricated IT pods, UPS, switchgear, cooling, racks, DCIM and services | Quote-based and oriented toward substantial projects; ecosystem dependence requires review |
| Vertiv | AI and HPC sites needing power and thermal infrastructure | UPS, switchgear, busway, batteries, modular systems, racks, CDUs and thermal management | Integration and service-contract scope should be examined carefully |
| Eaton | Electrical distribution, storage and microgrid projects | Medium-voltage gear, UPS, power distribution, BESS, microgrids and power management | May require partners for a complete liquid-cooling or IT-pod deployment |
No vendor should be selected merely because the vendor appears on a supply-chain map. A credible evaluation asks whether the vendor can document production capacity, meet the project schedule, support the target IT load, provide local service and parts, tolerate a second source, and include integration, commissioning, software, and maintenance in the quote.
What does resilience mean in a data center supply chain?
Supply-chain resilience has four layers: supplier resilience, logistics resilience, technical resilience, and operational resilience. A project needs multiple qualified manufacturers, alternative transport routes, interoperable equipment and spare capacity, and trained people who can maintain and replace the systems.
Inventory alone is not resilience. A spare component must be compatible, correctly stored, serviceable, supported by the manufacturer, and deployable by trained staff. A two-supplier strategy also fails when both suppliers depend on one sub-tier source.
Windows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallCrashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteThe most resilient designs minimize late-stage changes. They freeze interfaces early, qualify alternates before emergencies occur, divide capacity into independently usable phases, document factory and site tests, and preserve enough operational knowledge to maintain the installation after the construction team leaves.
Frequently Asked Questions
What is the biggest data center supply-chain bottleneck in 2026?
Power availability and grid interconnection are generally the highest-impact bottlenecks in 2026, followed by transformers, switchgear, high-density cooling, complete AI systems, and commissioning labor. The exact order varies by site, but a missing electrical or utility milestone can delay an entire facility.
Does having GPUs mean a data center has usable AI capacity?
No. GPUs or other accelerators are only one part of usable AI capacity. A working cluster also needs matching memory, host servers, networking, optics, racks, power distribution, cooling, software, and successful commissioning.
How long can large data-center transformers take to procure?
Large power transformers can take up to four years to procure, according to an IEA transmission-industry survey. The figure applies to large transformers and should not be generalized to every transformer used in a data center.
Is liquid cooling required for every AI data center?
No. Liquid cooling depends on rack density, accelerator type, workload, climate, facility design, and retrofit conditions. Direct-to-chip, immersion, rear-door heat exchangers, and hybrid air-liquid systems each create different plumbing, maintenance, water, and compatibility requirements.
What proves that an announced data center is actually deliverable?
The strongest evidence is a documented utility energization date, ordered long-lead electrical equipment, a funded and compatible IT fit-out, validated cooling and networking designs, completed commissioning tests, qualified operations staff, and customer-ready revenue service. An announcement or land purchase proves none of those milestones by itself.
The Bottom Line
Bottom line: The state of data center supply chains in 2026 is best described as capacity expansion under synchronized scarcity. The winners will not simply obtain more GPUs. They will coordinate utility power, transformers, electrical distribution, cooling, networking, construction, labor, commissioning, and operations with fewer late-stage design changes. For anyone evaluating a project, the decisive question is not “How many megawatts or accelerators have been announced?” but “Which independently verified milestones prove that usable, reliable capacity can enter service?”
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
