at Financial Technnology Year
Please contact them if you have any questions.
AWS HPC (High Performance Computing) from AWS (Amazon Web Services)
Offers a range of services to run HPC applications on-demand in the cloud, including parallel computing capabilities, scalability, and pay-as-you-go pricing.
More about AWS (Amazon Web Services)
Product analysis by function
High-Performance Computing Clusters for Actuarial Services
Hardware solutions that provide the computational power needed for complex actuarial simulations, Monte Carlo analyses, and other processor-intensive calculations required for modern pension management.
More High-Performance Computing Clusters
More Actuarial Services ...
Compute Performance
(5 Yes /5 Known /11 Possible features)
|
CPU Cores Total number of CPU cores available in the cluster. |
No information available | |
|
CPU Clock Speed Maximum clock speed of the CPUs in the cluster. |
No information available | |
|
GPU Availability Availability of GPUs optimized for parallel computations such as Monte Carlo simulations. |
AWS EC2 provides GPU-enabled instances (e.g., P4, G4, G5) commonly used in HPC and parallel actuarial computations. | |
|
GPU Cores Number of GPU compute cores for accelerated calculations. |
No information available | |
|
SIMD/Vector Capabilities Support for SIMD/vectorized instructions to speed up actuarial algorithms. |
AWS EC2 supports AVX, AVX-512, and other SIMD instruction sets in many instance families for vectorized operations. | |
|
Floating Point Operations Per Second (FLOPS) Peak floating point computation capacity of the cluster. |
No information available | |
|
Memory Bandwidth Maximum rate at which data can be transferred to/from memory. |
No information available | |
|
Low-Latency Interconnects Presence of high-speed, low-latency networks between nodes (e.g., InfiniBand). |
AWS offers Elastic Fabric Adapter (EFA) for low-latency, high-bandwidth interconnects comparable to InfiniBand. | |
|
Node Scalability Maximum number of nodes that can be integrated into the cluster. |
No information available | |
|
Elasticity Ability to dynamically add or remove computational resources based on demand. |
AWS Auto Scaling and parallel cluster services allow dynamic resource provisioning and elasticity. | |
|
Performance Benchmarks Availability of standardized performance benchmarks for actuarial workloads. |
AWS provides standardized HPC performance benchmarks and publishes case studies. |
Data Management
(6 Yes /6 Known /10 Possible features)
|
Total Storage Capacity Maximum storage available to the cluster. |
No information available | |
|
High Speed Storage Use of high-performance storage such as NVMe or SSD. |
AWS uses SSD (EBS gp3/io2) and NVMe for high speed storage options. | |
|
Data Redundancy Support for RAID or other data redundancy mechanisms. |
RAID support and EBS multi-attach can enable redundancy; S3 provides 99.999999999% durability for objects. | |
|
Backup Frequency How often automated backups are taken. |
No information available | |
|
Encryption At Rest Whether data is encrypted while stored. |
EBS, S3, and FSx data are encrypted at rest as a standard AWS feature. | |
|
Data Archiving Support for long-term, low-cost data archiving solutions. |
S3 and Glacier support data archiving for low-cost, long-term storage. | |
|
Data Compression Ability to compress data for more efficient storage. |
File systems like FSx and EBS offer data compression options. | |
|
Shared File Systems Availability of distributed file systems accessible by all compute nodes. |
FSx for Lustre, EFS, and shared file options exist for distributed access across compute nodes. | |
|
Automated Data Cleansing Built-in support for automated validation and cleansing of dataset inputs. |
No information available | |
|
Data Versioning Ability to maintain and roll back to previous versions of datasets. |
No information available |
Job/Workload Management
(10 Yes /10 Known /10 Possible features)
|
Job Scheduler Existence of advanced job scheduling system (e.g., SLURM, PBS, LSF). |
HPC users can use AWS ParallelCluster, which supports schedulers including SLURM, AWS Batch. | |
|
Job Prioritization Ability to prioritize critical actuarial jobs according to business rules. |
Most job schedulers (e.g., SLURM) on AWS support job prioritization. | |
|
Resource Quotas Support for allocating resource quotas per user or project. |
Resource quotas can be set per user/project via scheduler and AWS account management tools. | |
|
Job Monitoring Real-time monitoring of running jobs and their resource usage. |
ParallelCluster and supported schedulers provide real-time job/resource monitoring. | |
|
Fault Tolerance Ability to automatically recover or restart failed jobs. |
Schedulers support job checkpointing and re-queuing failed jobs automatically. | |
|
Automated Notifications Automatic alerts for job completion, failures, or resource exhaustion. |
AWS Batch and schedulers provide automated email/SNS notifications for job completion/failure. | |
|
Workflow Automation Support for automated, multi-stage actuarial workflows. |
AWS Batch, Lambda, and workflow engines allow automation of multi-stage workflows. | |
|
API for Job Submission REST or command-line interface for automated job submissions. |
AWS Batch, scheduler REST/CLI APIs enable programmatic job submission. | |
|
Job Array Support Ability to efficiently run large arrays of similar actuarial jobs. |
SLURM on AWS ParallelCluster and AWS Batch both support job arrays. | |
|
Historical Job Logs Access to detailed history and logs of previous jobs for audit purposes. |
Job logs/history stored in CloudWatch, S3, and/or within schedulers. |
Software Compatibility
(10 Yes /10 Known /10 Possible features)
|
Support for Actuarial Software Compatibility with common actuarial software (Prophet, MoSes, AXIS, etc.). |
AWS provides AMIs for actuarial software (Prophet, R, MATLAB, Python, etc.) and supports custom installation. | |
|
Statistical Programming Languages Availability of R, Python, MATLAB, and related packages. |
AWS supports R, Python, MATLAB, and other statistical tools as AMIs and via AWS Marketplace. | |
|
Custom Model Integration Ability to deploy custom-built simulation and projection models. |
Custom Docker containers and AMIs allow deployment of proprietary/custom models. | |
|
High-Performance Libraries Pre-installed numerical and actuarial libraries (BLAS, LAPACK, TensorFlow, etc.). |
Many scientific and numerical libraries available as AMIs or containers on AWS. | |
|
Containerization Support Ability to run software in Docker, Singularity, or other container platforms. |
AWS supports Docker and Singularity for containerized workloads. | |
|
Operating System Support Support for preferred OS (Linux, Windows, etc.). |
Both Windows and Linux supported for compute nodes. | |
|
Parallel Computing Interfaces Support for MPI, OpenMP, or equivalent parallel computing frameworks. |
Parallel computing (MPI, OpenMP) supported via AWS ParallelCluster, EC2 images, and AMI selection. | |
|
Cloud Integration Ability to burst workloads or integrate with public/private clouds. |
AWS enables hybrid cloud and bursting to public AWS Cloud. | |
|
License Management Facilities for monitoring and managing proprietary software licenses. |
Tools exist for tracking license usage (e.g. FlexNet, AWS License Manager) in enterprise environments. | |
|
Version Control Integration Easy integration with Git or similar version control systems for code management. |
CodeCommit and other services enable integration with Git and other VCS. |
Security and Compliance
(9 Yes /9 Known /10 Possible features)
|
User Authentication Multi-factor authentication and single sign-on for user access. |
AWS supports Multi-Factor Authentication (MFA) and SSO via IAM for secure access. | |
|
Role-Based Access Control Granular assignment of permissions and roles to users. |
IAM allows granular role and permission definition for all cloud resources. | |
|
Encryption in Transit Data and communications encrypted over the network. |
Encryption in transit is default (TLS/SSL) for AWS APIs and internal communications. | |
|
Audit Trails Logged record of user actions for compliance and investigation. |
AWS CloudTrail provides comprehensive audit logs for actions in the environment. | |
|
GDPR/Local Compliance Support for compliance with data protection regulations such as GDPR. |
AWS offers GDPR and other regional compliance certifications. | |
|
Vulnerability Management Regular patching and vulnerability scanning of hardware and software. |
AWS performs regular vulnerability scanning and patching as part of its security model. | |
|
Disaster Recovery Comprehensive disaster recovery and data restore procedures. |
AWS Backup, multi-region replication, and documented disaster recovery processes are offered. | |
|
Physical Security Physical security measures for on-premise cluster deployments. |
No information available | |
|
Data Masking Tools to anonymize or mask sensitive pension fund data. |
Data masking/anonymization available through partner solutions and AWS services (e.g., Macie). | |
|
Endpoint Protection Malware and intrusion detection for all endpoints in the cluster. |
AWS GuardDuty, Inspector, and endpoint security tools provide malware detection. |
Scalability and Flexibility
(10 Yes /10 Known /10 Possible features)
|
Horizontal Scalability Ability to add computing nodes with minimal configuration changes. |
Cluster/node scaling is configurable and can add nodes dynamically with minimal setup. | |
|
Vertical Scalability Ability to increase the capacity (CPU, RAM) of existing nodes. |
Instance type upgrades can be performed for compute/memory scaling. | |
|
Resource Pooling Dynamic allocation of shared resources between projects or departments. |
Resource sharing can be administratively managed across users/projects in the same AWS account. | |
|
Custom Configuration Support for custom node types or heterogeneous clusters. |
Cluster nodes can be custom-defined (ARM/x86, GPU/CPU mix, etc.). | |
|
Bursting to Cloud Capacity to support hybrid on-premises and cloud configurations. |
Hybrid cloud architecture (on-prem + AWS) supported through AWS Outposts and Direct Connect. | |
|
Self-Service Resource Management Users can request or release resources without admin intervention. |
Users can provision resources via the console or APIs without admin intervention. | |
|
Elastic Storage Dynamically assign storage as project data grows. |
Storage resources dynamically scaled using EFS/FSx/S3 as needed. | |
|
Multi-Tenancy Ability to securely isolate environments for different teams. |
Logical multi-tenancy offered via AWS accounts, VPC, and IAM. | |
|
Resource Quota Management Set limits on usage for predictable cost and capacity management. |
Resource limits, quotas, and auto-scaling define resource usage and cost. | |
|
Automated Provisioning Automated setup and teardown of computational resources as needed. |
CloudFormation, ParallelCluster, and APIs allow automation of provisioning/teardown. |
Reliability and uptime
(7 Yes /7 Known /10 Possible features)
|
Uptime SLA Guaranteed percentage uptime by vendor/service provider. |
No information available | |
|
Redundant Power Supply Multiple power sources to prevent cluster outages. |
AWS data centers and hardware are built for redundancy. | |
|
HA Clustering Support for high availability clustering and failover. |
High availability clustering architectures and managed database failover available. | |
|
Hot Swappable Components Ability to replace or upgrade hardware without shutting down. |
No information available | |
|
Automated Health Monitoring Real-time monitoring and alerts for hardware or software failures. |
AWS CloudWatch and Health provide proactive infrastructure/system monitoring. | |
|
Automatic Node Recovery Automatic reboot or recovery of failed nodes. |
Automatic recovery/reboot features available for EC2 and cluster nodes. | |
|
Service-Level Monitoring Continuous monitoring for critical actuarial services. |
CloudWatch Synthetics and other tools enable continuous application/service-level monitoring. | |
|
Spare Node Capacity Built-in spare nodes for immediate failover. |
No information available | |
|
Scheduled Maintenance Windows Clearly defined and communicated downtime for upgrades/maintenance. |
Maintenance windows scheduled and documented; communicated via AWS Service Health Dashboard. | |
|
Error Correction Codes (ECC) Memory RAM with ECC to protect against data corruption. |
Many AWS instance types provide ECC memory. |
User Experience and Accessibility
(9 Yes /9 Known /10 Possible features)
|
Web Portal Access User-friendly web interface for accessing and managing cluster resources. |
AWS Console provides a user-friendly web portal for managing clusters/resources. | |
|
Command-Line Utilities Robust CLI for advanced users and automation. |
AWS CLI and SDKs are robust, well-documented, and extensively used. | |
|
Multi-Language Support Documentation and interface availability in multiple languages. |
AWS documentation supports multiple languages for global accessibility. | |
|
Accessibility Features Compliance with accessibility standards for users with disabilities. |
No information available | |
|
Self-Service Documentation Comprehensive knowledge base and troubleshooting guides. |
AWS provides comprehensive online documentation, knowledge bases, and troubleshooting resources. | |
|
Collaboration Tools Integration Integration with email, chat, and documentation platforms. |
AWS services can integrate with Slack, Teams, email, and other collaborative tools. | |
|
Single Sign-On Unified login experience across platforms and tools. |
AWS SSO and identity federation allow single sign-on. | |
|
Customizable Dashboards User dashboards that can be configured to display relevant information. |
CloudWatch dashboards are customizable and user-specific. | |
|
Mobile Access Support for monitoring or interacting with the system from mobile devices. |
AWS Management Console supports web/mobile access and CloudWatch mobile app is available. | |
|
API Documentation Accessible and well-maintained documentation for all system APIs. |
API Gateway and service documentation are comprehensive and well-maintained. |
Integration and Interoperability
(10 Yes /10 Known /10 Possible features)
|
RESTful API Provision of standards-compliant APIs for external integrations. |
RESTful API access is standard for AWS services including batch, S3, EC2. | |
|
Database Connectivity Ability to connect to internal and external databases securely. |
All major database types available (RDS, Redshift, Aurora); secure connections supported. | |
|
File Format Compatibility Support for a range of input/output formats (CSV, Excel, Parquet, HDF5, etc.). |
Support for CSV, Parquet, JSON, Excel, HDF5 through S3 and Glue ETL/DataBrew. | |
|
ERP/CRM Integration Integration with enterprise pension, HR, and financial systems. |
Integration available via AWS Marketplace and multiple third-party ERP/CRM connectors. | |
|
Data Lake Integration Support for ingesting and exporting data from/to data lakes. |
Glue, Athena, and Lake Formation natively support data lake integrations. | |
|
Event-Driven Architecture Support for pub/sub messaging or webhooks for real-time data flow. |
SNS, SQS, EventBridge, Lambda all support real-time, event-driven architectures. | |
|
Authentication Federation Integration with enterprise identity management (LDAP, SAML, etc.). |
Integration with LDAP, SAML, and other enterprise authentication methods via AWS SSO and Directory Service. | |
|
Orchestration Integration Compatibility with orchestration tools like Kubernetes or Airflow. |
Kubernetes (EKS), ECS, Batch, and Airflow are all available as managed services. | |
|
Partner Data Services APIs or connectors for industry-standard actuarial and fund management services. |
AWS Data Exchange, Marketplace, and partner APIs support industry-standard integrations. | |
|
Legacy System Support Interoperability with older or proprietary pension management platforms. |
Legacy system support available through AWS Migration Hub, DataSync, and partner connectors. |
Cost Management and Efficiency
(10 Yes /10 Known /10 Possible features)
|
Usage-Based Billing Ability to measure and bill based on actual resource usage. |
Billing by instance-second (EC2) and by resource usage is fundamental to AWS pricing. | |
|
Cost Reporting Detailed reporting of costs by department, project, or user. |
Cost Explorer and tagging facilitate detailed reporting by account/project/user. | |
|
Energy Efficiency Use of hardware and cooling solutions to minimize energy costs. |
AWS Graviton, right-sizing, and energy efficient data centers support energy efficiency. | |
|
Idle Resource Detection Automated identification and deallocation of idle resources. |
AWS Compute Optimizer and auto-scale settings enable idle resource detection/deallocation. | |
|
Budget Alerting Automated notifications when approaching or exceeding budget thresholds. |
Budget alerts can be set up in AWS Budgets with email/SNS notifications. | |
|
Capacity Planning Tools Forecasting future computational needs and hardware investments. |
AWS Compute Optimizer and integrated dashboard supports capacity planning. | |
|
Spot/Preemptible Resource Support Access to discounted, interruptible compute resources where possible. |
Spot/preemptible instance types (EC2 Spot) are a major AWS offering. | |
|
Procurement Integration Integration with financial systems for asset tracking and procurement planning. |
AWS Marketplace/API integrations with ERP/Finance product for procurement tracking. | |
|
Chargeback/Showback Features Reporting tools for allocating costs to appropriate business units. |
Showback/chargeback reporting supported with AWS Cost Allocation Tags. | |
|
Power Usage Metrics Real-time and historical metrics on power consumption. |
AWS provides via CloudWatch and service-specific dashboards real-time and historical power usage metrics. |
Support and Vendor Services
(10 Yes /10 Known /10 Possible features)
|
24/7 Support Availability Access to technical support at all times. |
AWS Business and Enterprise support provides 24/7 access. | |
|
Dedicated Account Manager Assigned manager for ongoing relationship and escalations. |
Dedicated AWS account managers are available for Enterprise support tiers. | |
|
SLAs for Incident Response Service-level guarantees for support ticket response and resolution times. |
Support contract defines SLAs for ticket response and escalation. | |
|
Proactive System Monitoring Vendor provides monitoring and alerts on infrastructure health. |
Proactive health monitoring and alerts are core part of AWS support contracts. | |
|
Training and Onboarding Structured training and knowledge-transfer sessions for actuarial teams. |
AWS offers free and paid onboarding/training for customers and partners. | |
|
Professional Services Availability of consultants for system customization/integration. |
AWS Professional Services offers consulting, system integration, and custom engagements. | |
|
Hardware Replacement SLAs Guaranteed time frames for hardware repair or replacement. |
Enterprise support includes hardware replacement SLAs in case of direct AWS hardware use. | |
|
Knowledge Base Access Comprehensive, searchable knowledge base of issues and solutions. |
AWS knowledge base is comprehensive and searchable. | |
|
Community Forums Access to vendor-supported discussion forums. |
AWS hosts active community forums and developer communities. | |
|
Roadmap Transparency Visibility into vendor's future feature and upgrade plans. |
AWS public roadmap available and communicated transparently. |
This data was generated by an AI system. Please check
with the supplier. More here
While you are talking to them, please let them know that they need to update their entry.