Purpose-built Solutions for Networking
Can't find the solution you are looking for?
Contact us to explore a design partnership opportunity!
Modern Data Centers
Modern Data Centers are the backbone for many critical workloads at large enterprises, yet traditional tools for monitoring network performance are manual and reactive.
SliceUp proactively detects known and zero-day issues so that the Ops Teams can fix them before the end-user experience is degraded. The types of problems SliceUp detects include:
* Syslog ML - unknown unknowns - looking for long tail issues, including system errors, configuration errors, network outages, unauthorized access attempts, and service disruptions
* Latency - analyzing round trip and retransmission delay to get ahead of issues that may impact how the user is experiencing the network
* Interface Utilization - looking for early indicators of whether an interface will be over-utilized in the near future
* Fiber Channel - looking for Latency and Interface Utilization issues in Fiber Channel Network
* Hardware Lifecycle - proactively address potential hardware failures before they occur and minimize downtime and costs associated with emergency repairs or replacements
* Exchange Logs - looking for performance issues that may include slow email delivery times, database mounting delays, high latency in client-server communication
* VoIP - proactively monitoring EMOS scores, jitter, packet loss, and delays for quality assurance of your voice and video calls
* Digital Twin - our "what if calculator" models different future scenarios to identify bottlenecks, adjusted for risk tolerance. Examples of metrics measured are Interface Utilization, TCAM, and Memory
* Auto Remediation - system scans the streaming syslog data for matching anomalous log templates (issues) and triggers the automated workflow that executes the Ansible script
* More coming soon
HPC/GPU Data Centers
HPC/GPU Data Centers are becoming popular with the rise of GenAI (Large Language Models) and simulation technologies, which require ultra-low latency and high throughput.
SliceUp enables operators to optimize the time it takes to train Large Language Models by delivering end-to-end visibility to maximize utilization and minimize congestion. The types of problems SliceUp is working on include:
* Syslog ML
* Per hop Latency
* Utilization
* Congestion
* Traffic Polarization
* Flow completion times, and more
WAN / SD-WAN
Ensuring high performance across a distributed WAN network and managing bandwidth demands due to increasing data volumes and applications is business critical.
SliceUp proactively detects known and zero-day issues so that the Ops Teams can fix them before they affect the end-user experience. The types of problems SliceUp detects include:
* Syslog ML - unknown unknowns - looking for long tail issues, including system errors, configuration errors, network outages, unauthorized access attempts, and service disruptions
* Path change - SliceUp is alerting where network path adjustments and delays have occurred even when the route or endpoint isnʼt directly under your control
* Neighbor Drops (BGP, IS-IS)
* VoIP - proactively monitoring EMOS scores, jitter, packet loss, and delays for quality assurance of your voice and video calls
* Digital Twin - our "what if calculator" models different future scenarios to identify bottlenecks and adjust them for risk tolerance. Examples of metrics measured are Interface Utilization, TCAM, and Memory
* Auto Remediation - The system scans the streaming syslog data for matching anomalous log templates (issues) and triggers the automated workflow that executes the Ansible script
* More coming soon
An example use case - an unusual spike in retransmission delay caused by duplex mismatch, fixed with automated remediation workflow.
Digital Twin for “What If Analysis”
Auto-discovered Network Topology. SliceUp integrates with existing SNMP data collection systems to auto-discover your network topology. By overlaying key metrics on an always current topology, network administrators can easily find where the future problems are.
Plan Ahead - Get Proactive: SliceUp allows you to withstand the next big product launch or a black swan event. Proactively mitigate the risk based on modeled scenarios and take the appropriate action - add additional links, increase bandwidth, or re-route traffic.
Identify Resources Needed. SliceUp's "what if calculator" models different future scenarios to identify bottle necks and see which interfaces may become over utilized if traffic increases, adjusted for risk. Examples of metrics measured: Interface Utilization, TCAM, Memory.
Can't find the solution you need?
FAQ:
How is it deployed? SliceUp is a Kubernetes-based, scalable deployment that can be installed in your Private Cloud or On-Prem. This way, your data never has to leave your environment. With our fully automated data preparation and parsing, it is designed to work well even in an air-gapped environment.
What are your existing out-of-the-box integrations? Some of the existing integrations include Syslog, Kafka, NetScout, Ansible, ServiceNow, SolarWinds, NNMI, and others. Contact us for a full list of integrations, or to discuss a custom integration to meet your needs.
How much historical data do I need for it to work? The system needs a minimum of a week's worth of historical data, but we recommend 30 days’ worth of data for it to establish the baseline for your environment accurately.
