AI Databases & Vector Stores
How do I contact ApertureData sales for the Custom/Enterprise plan and what info should I bring (data size, QPS, modalities, deployment)?
7 min read
Most teams exploring ApertureDB’s Custom/Enterprise plan already feel the pain of multimodal AI at scale—billions of metadata entries, tight latency budgets, and real SLAs. To move quickly to the right sizing and architecture, it helps to come into the sales conversation with a clear picture of your data size, QPS, modalities, and deployment constraints.
Quick Answer: Contact ApertureData sales for the Custom/Enterprise plan through the contact form at https://www.aperturedata.io/contact-us or by emailing team@aperturedata.io. For a useful first conversation, bring approximate data volume (media, metadata, embeddings), target QPS and latency, your modalities (text, images, video, audio, documents), and deployment preferences (cloud, VPC, on‑prem).
Frequently Asked Questions
How do I contact ApertureData sales for the Custom/Enterprise plan?
Short Answer: Use the contact form at https://www.aperturedata.io/contact-us or email team@aperturedata.io and indicate that you’re interested in the Custom/Enterprise plan for ApertureDB.
Expanded Explanation:
For Custom/Enterprise deployments, we typically start with a short discovery call focused on your workload: multimodal RAG, GraphRAG, agent memory, or dataset preparation at scale. The fastest way to initiate this is through the contact form on our site—choose a relevant interest (e.g., “Enterprise deployment,” “Custom quote,” or similar) and briefly describe your data scale and latency expectations.
If you prefer, you can also reach out directly via email at team@aperturedata.io. Provide a one‑paragraph summary of your use case—modalities, expected data growth, and where you plan to run (AWS/GCP/VPC/on‑prem). From there, our team will suggest a call to size the deployment, discuss SLAs, and map you to either ApertureDB Cloud or a dedicated/VPC/on‑premise setup.
Key Takeaways:
- Use the contact form or email team@aperturedata.io, and mention you want Custom/Enterprise.
- A short, concrete description of your workload accelerates sizing, pricing, and architecture recommendations.
What information should I prepare before speaking with ApertureData sales?
Short Answer: Come prepared with estimates for your data size, QPS/latency targets, modalities, and deployment preferences, plus a brief description of how you’ll use ApertureDB (RAG, GraphRAG, agents, dataset prep).
Expanded Explanation:
ApertureDB is a foundational data layer—vector + graph + multimodal storage—so the “right” plan depends on how much data you’ll store (media, metadata, embeddings), how fast you need to query it (QPS and tail latency), and where you need it deployed (cloud, VPC, on‑prem). The more specific you can be, the less time you’ll spend iterating on back‑of‑the‑envelope guesses.
You don’t need exact counts; reliable ballpark numbers are enough. We expect estimates to change as you prototype. We use your inputs to choose the right architecture (cluster size, replicas, storage class) that can hit sub‑10ms vector search, ~15ms graph lookups at your target scale, without forcing you into fragile pipelines or over‑provisioned infrastructure.
Steps:
- Summarize your use case: e.g., “Multimodal RAG over PDFs + images,” “GraphRAG for product catalog + manuals,” or “Agent memory for customer support across chat + call transcripts.”
- Estimate scale and traffic: total records, embeddings, media size, expected QPS, acceptable latency, and growth over 12–24 months.
- Clarify deployment & security requirements: preferred cloud, VPC vs on‑prem, RBAC/SOC2/SLA needs, and any data residency constraints.
How detailed should my data size and QPS estimates be for Custom/Enterprise planning?
Short Answer: Rough orders of magnitude are enough (e.g., 10M vs 100M documents; 1K vs 10K QPS), but you should be clear on peak QPS, latency expectations, and projected growth.
Expanded Explanation:
When you’re evaluating a foundational data layer like ApertureDB, we’re designing for growth and stability, not just a demo. We don’t need you to know you’ll have exactly 247,382,119 embeddings; we do need to understand whether you’re closer to 10M or 1B+ records and whether you expect 500 vs 15,000 queries per second.
For Custom/Enterprise plans, we’re typically optimizing across three axes:
- Capacity: media (images, videos, audio, documents), embeddings, and metadata (often 1.3B+ entries).
- Throughput: queries per second—KNN queries, GraphRAG traversals, and metadata‑heavy filters.
- Latency: p95/p99 targets for vector search and graph lookups (sub‑10ms vector search, ~15ms graph lookups are realistic at scale with the right sizing).
Bringing even rough numbers on each axis lets us design an architecture that can deliver 2–10X faster KNN and 10K+ QPS without instability—so your team is asleep at 5AM, not babysitting infrastructure.
Comparison Snapshot:
- Option A: Hand‑wavy estimates – “We’ll have a lot of data and need it to be fast.”
- Option B: Order‑of‑magnitude estimates – “We expect 50–100M documents, 200–300M embeddings, 5–8K QPS, sub‑50ms p95.”
- Best for: Option B is best; it gives us enough signal to size clusters, replicas, and storage strategy without wasting cycles.
What modalities and deployment details should I share for a Custom/Enterprise quote?
Short Answer: List all modalities you care about (text, images, videos, audio, PDFs/Office docs, annotations/bounding boxes) plus your deployment constraints (AWS/GCP/VPC/on‑prem, RBAC, SOC2, data residency) so we can recommend the right ApertureDB setup.
Expanded Explanation:
ApertureDB is multimodal‑native. We store and query images, videos, audio, documents, and text natively alongside embeddings and metadata in one database—no conversion, no fragmentation. For Custom/Enterprise plans, knowing which modalities matter (and how they relate) lets us design the right vector + graph footprint and storage tiering.
Deployment details are equally important. Some teams run fully in ApertureDB Cloud; others need an isolated VPC or on‑prem installation due to compliance, data residency, or latency to adjacent systems. Enterprise customers often require RBAC, SSL everywhere, SOC2 posture, and sometimes pen‑test evidence. Sharing those requirements upfront prevents surprises later and directly informs which SKUs and SLA tiers make sense.
What You Need:
- Modality breakdown:
- Text (chat logs, docs, code), PDFs/Office documents
- Images (product photos, frames from video)
- Videos (full length, clips), audio (calls, podcasts)
- Annotations/bounding boxes, labels, and other metadata
- Deployment & security expectations:
- Preferred environment: ApertureDB Cloud, AWS/GCP, VPC, Docker/K8s, or fully on‑prem
- Security/compliance: SOC2, SSL, RBAC, data residency, backup/DR requirements
How does this information influence the Custom/Enterprise plan and overall strategy?
Short Answer: Your data size, QPS, modalities, and deployment constraints drive how we architect your ApertureDB cluster, what SLA and pricing tier we recommend, and how quickly you can go from prototype to production.
Expanded Explanation:
Custom/Enterprise isn’t “bigger license, same architecture.” It’s about matching the foundational data layer to your actual workloads so you get predictable performance and TCO, not surprise re‑architecture six months in. When we understand your multimodal mix, graph complexity, and traffic patterns, we can:
- Tune vector search (distance metrics, index layouts, sharding) to hit sub‑10ms search at your scale.
- Shape the property graph to support GraphRAG and agent memory with ~15ms lookups on billion‑scale graphs.
- Align replicas, high‑availability setup, and backup strategy with your uptime and RPO/RTO requirements.
- Use ApertureDB Cloud workflows (Ingest Dataset, Generate Embeddings, Detect Faces and Objects, Direct Jupyter Notebook Access) to compress your setup time—saving 6–9 months of infrastructure work and taking you from prototype → production up to 10× faster.
The outcome is not just “faster queries.” It’s a unified multimodal memory layer where vectors, metadata, and relationships live together—no fragile pipelines, no constant schema rewrites—so your RAG, GraphRAG, and agents can search with context, not just similarity.
Why It Matters:
- Lower risk, higher stability: Right‑sized architecture avoids the 4K QPS “works in dev, falls over in prod” failure mode and gets you to 10K+ QPS with a high degree of stability.
- Faster time to value: A well‑scoped Custom/Enterprise plan, grounded in your real workload, means less integration work and more time building intelligent agents and applications on top of a solid data foundation.
Quick Recap
To contact ApertureData sales for the Custom/Enterprise plan, start with the contact form at https://www.aperturedata.io/contact-us or email team@aperturedata.io. For a high‑value first conversation, bring rough estimates of your dataset size (media, metadata, embeddings), expected QPS and latency targets, the modalities you care about (text, images, video, audio, documents, annotations), and your deployment/security constraints (cloud vs VPC vs on‑prem, RBAC, SOC2, data residency). These inputs let us design the right vector + graph + multimodal storage architecture, map you to the correct Custom/Enterprise tier, and help you move from prototype to production 10× faster with predictable performance and TCO.