Compute Models
Instances, containers and functions, and what each is priced and shaped for.
5 to work through
-
intermediate
A fleet of 200 instances averages 15% CPU and 85% memory. Finance wants a 40% cost reduction. What do you do?
2 min answer -
intermediate
A workload runs continuously at moderate load. Finance asks why the serverless bill is four times the previous instance cost. Explain the crossover and how you would calculate it.
2 min answer -
advanced
A GPU platform has a few very large jobs competing with thousands of small ones. How should scheduling fairness, fragmentation, bin packing, preemption, queueing and starvation shape the architecture?
2 min answer -
advanced
A compute platform schedules GPU jobs with execution times ranging from minutes to weeks, where large jobs need many co-located accelerators. What happens if the scheduler uses simple first-come-first-served with best-fit packing?
2 min answer -
advanced
A device-testing platform receives a burst of tens of thousands of test sessions after a major framework release. How should queues, worker pools, concurrency limits, resource classes, scheduling fairness and backpressure prevent capacity exhaustion?
2 min answer
5 terms in this topic
Burstable Instance
An instance that provides a low baseline CPU allocation and accrues credits while idle, spendable for short periods of full performance.
conceptCompute Models
Virtual machines, containers, serverless functions and dedicated hardware — chosen by workload shape rather than by modernity.
conceptFragmentation Cost
The capacity that exists in aggregate but cannot be used because it is scattered across nodes in pieces too small for the jobs that need it - the dom…
conceptGang Scheduling
Allocating every resource a job needs simultaneously or not at all, so tightly-coupled distributed work cannot deadlock holding a partial allocation.
conceptInstance Family
A group of instance types sharing a resource profile — general purpose, compute optimised, memory optimised, storage optimised, accelerated — chosen …
Neighbouring topics
Cloud Architecture
General material on designing for cloud platforms.
Cloud Storage
Object, block and file storage, and the access patterns each suits.
Cloud Databases
Managed relational, key-value, document and analytical services.
Containers
Images, registries, immutability and the deployment model they enable.
Kubernetes
The reconciliation loop, and whether the workload needs what it provides.
Serverless
Scale to zero, per-request billing, cold starts and connection limits.
Autoscaling
Signals, delays and bounds — and the maximum that caps a runaway bill.
Load Balancing
Distributing traffic, health checking, and removing failures from rotation.
Multi-Region Architecture
Surviving a region, and the data consistency price of doing so.
Availability Zones
The unit of correlated physical failure, and what zones do not protect against.
Disaster Recovery
Backup-restore, pilot light, warm standby and active-active postures.
Backup Strategies
Scope, immutability, separation, and the restore drill that makes it real.
Infrastructure as Code
Declarative infrastructure, drift, state files and rebuild-from-empty.
Landing Zones
A governed foundation of accounts, network, identity and guardrails.
Managed Services
Which operational responsibilities actually transfer, and which do not.
Cloud Migration
Per-application disposition, sequencing and the capability change underneath.
Multi-Cloud
Best-of-breed, portfolio and portable — three very different costs.
Edge Computing
Moving compute towards the user, and what cannot follow it.
Cloud Governance
Preventive policy, tagging, quotas and cost and security guardrails.